(Topic ID: 284486)

WPC - Stitching a through hole - advice sort

By Biglouie

3 years ago


Topic Heartbeat

Topic Stats

  • 51 posts
  • 8 Pinsiders participating
  • Latest reply 3 years ago by Biglouie
  • No one calls this topic a favorite

You

Linked Games

Topic Gallery

View topic image gallery

IMG_20201228_200224 (resized).jpg
IMG_20201228_193942 (resized).jpg
IMG_20201228_191807 (resized).jpg
address A0 (resized).jpg
IMG_20201228_173711 (resized).jpg
IMG_20201228_172745 (resized).jpg
IMG_20201228_172755 (resized).jpg
IMG_20201227_211839 (resized).jpg
IMG_20201227_231421 (resized).jpg
04_U08_local.jpg
03_U08_address_bus1.jpg
02_U08_address_bus0.jpg
01_U08_data_bus.jpg
IMG_20201227_205055 (resized).jpg
IMG_20201227_204248 (resized).jpg
IMG_20201227_185848 (resized).jpg

There are 51 posts in this topic. You are on page 1 of 2.
#1 3 years ago

Hi all, Merry Christmas.

I wonder if I could ask for some advice. I'm planning to add L8 to my #bop for my restoration project, so I need to jumper. It looks like it's seen some action before. Middle of the pic.

IMG_20201226_141331 (resized).jpgIMG_20201226_141331 (resized).jpg

There is a small amount on continuity where the trace meets the hole.

Do I gently sand back on the trace and pop a stitch in and solder it?

Or do I not sand it, and just try and get the stitch to make that small bit of connection?

First time doing a stitch. I've done some testing on a dud board, but wondering about the sanding.

'tuther side below for fun!
Screenshot_20201223-232509 (resized).jpgScreenshot_20201223-232509 (resized).jpg

Thanks in advance.

#2 3 years ago

Gave it a little sand.

Stitched and installed the jumper, all good.

IMG_20201226_154226 (resized).jpgIMG_20201226_154226 (resized).jpg

After that I installed nvram. Now it won't boot, d19 and d21 both constantly on.

Have put resistor r93 in and w3 is removed.

I've buzzed all the socket connections front and back. No joy. Feel like I took too much on, but I thought I'd done a lovely job as well.

IMG_20201226_192124 (resized).jpgIMG_20201226_192124 (resized).jpg
#3 3 years ago

Did you install the socket for the memory. Buzz it out top and bottom. I’m betting you either didn’t get good flow (you should always hit the holes with fresh solder and clean them out again) or killed another thru hole on the memory socket

#4 3 years ago

Buzzed out nvram socket, found 2 bad at pins 10 and 20

10 goes to CPU and 20 goes to ASIC. Both now buzzing after reflow.

All others buzz.

But I'm not getting anything on boot. I'll buzz them all again, see of anything comes up.

Could I have caused more damage by booting up with these two connections not in place?

Has to be connected to the nvram, as was working before I did this. Resistor is correct as far as I know, would that matter?

#5 3 years ago

Buzzed them all, top and bottom of board now, everything checks out

I can't imagine both nvram would be dead, so I'll keep digging..

#6 3 years ago

It’s socketed. Drop the original back in and see if it boots

#7 3 years ago

Hi Tallon, thanks for coming back to me.

It's socket as of this afternoon. The old 62256 ram had to be desoldered and it didn't make it out with all its legs. I had to cut a few as the solder sucker didn't manage to get everything out.

I don't have another ram to check with at the moment.

I must be missing something I'll try and find some new ram and give it a go.

I just went through all the pins top and bottom again and can buzz everything..

The dip switches wouldn't make a difference would they?

Could I have fried the ram if there were missing connections?

#8 3 years ago

Click on the picture from zaza and test out the connections from the U8 to everywhere these pins connect to.
https://www.pinwiki.com/wiki/index.php?title=File:WPC89_CPU_compnt-side_ZAZ_FEB2017-10.jpg

#9 3 years ago

The picture was super helpful, I used my schematics in conjunction with that to test them all.

Beeps everywhere..

If D19 and D21 are constantly on does that mean the CPU can't.see the ram or could that be any number of things.

I'll try without ribbons and maybe swap the CPU for a known worker.

Checked R93 resistor again, all good.

It's going to be a trace, I just know it, but it all buzzes so...

#10 3 years ago

CPU board with J210 only, constant d19 and d21

Tried two nvram - no joy
Buzz tested u8 to everywhere
Tried working CPU - no joy
Tried original, smaller size game ROM - no joy
Tried Hurricane game ROM as it's larger size - No joy
Pressed about a bit on chips - no joy
DIPs - 1 & 2 off, the rest all on
Checked R93 resistor incase that was not reading new 6264 ram as had 62259 before
W3 is removed

Tested some voltages on driver board, seems normal. 5v on driver is 4.7v tho at tp2, but all this was fine til nvram went in, so doubtful voltages have changed during this process?

If I had another ram, that would be another test. EDIT: HURRICANE ram is socketed. Tried nvram in there, no problems. So not the ram.

Back to pinwiki for more reading and "wpc d19 d21 stuck on nvram" Google searches

I haven't desoldered the old battery holder, that doesn't need to come off if nvram is there surely?
IMG_20201227_101908 (resized).jpgIMG_20201227_101908 (resized).jpg

I also notice from my reference pictures, when I got the machine the DIP settings were as below. From what I have read, 1 and 2 are always meant to be off?
Screenshot_20201227-104309 (resized).jpgScreenshot_20201227-104309 (resized).jpg

Both set ON in that picture.

#11 3 years ago
Quoted from Biglouie:

is 4.7v tho at tp2

You can measure voltage across C31 on the CPU but it will be close to 4.69 then... which is too low.

#12 3 years ago

Right, hang on. I thought you meant pin 31 on the cpu. That's becoase I numbered it like that in my note book for.buzz testing. Ha, I'm a fool. Thanks for coming back to me

Pin 31 reports 1.9v on CPU for what it's worth.

I'll go back out and test c31

Rob

#13 3 years ago

C32 (edit: I mean c31) is 4.72v

Black on gnd and red on +ive leg of c31

TP2 is 4.8v

On the game ROM, I measured across pin16 (gnd) and pin32 (vcc) and got 4.78v

#14 3 years ago
Quoted from Biglouie:

I haven't desoldered the old battery holder, that doesn't need to come off if nvram is there surely?

The point of installing NVRAM is to remove any possibility that alkaline batteries get reinstalled in the future by anyone. I always remove battery holder.

#15 3 years ago
Quoted from PinballManiac40:

The point of installing NVRAM is to remove any possibility that alkaline batteries get reinstalled in the future by anyone. I always remove battery holders.

Sure thing, but do you HAVE to remove them. Or I haven't yet, but it isn't causing this issue?

I'll pull it later anyway, but...

#16 3 years ago

Plus, you show on one picture with the batteries installed, a pretty good amount af corrosion is on the holder terminals. It is nearly impossible to clean off the corrosion with it installed. If you leave any trace of corrosion, it will find it's way to the board over time.

I will want to see a picture of the board with the folder removed. I saw some suspected areas of corrosion on resistors. I am hoping that is just because the camera is out of focus.

#17 3 years ago

so.....It worked before you worked on it and now it wont boot. always check your work. Make sure you didn't get a solder bridge under the socket. If need be, you can try removing the socket and hard soldering the nvram in. I would NOT recommend it! Remove the battery holder and get rid of it if your gonna use NVram. BTW, BOP is one if the games that has easter eggs tied to the clock.

https://pinside.com/pinball/forum/topic/repair-log-ij-cpu-d19-d21-locked-on

The other thing to keep in mind is pulling the board in and out can cause a corroded connection to finally break and leave you with a dead board. With corrosion showing on the holder, you could be chasing your tail trying to find a problem that else where on the board.

#18 3 years ago
Quoted from Biglouie:

C32 is 4.72v

Black on gnd and red on +ive leg of c32

TP2 is 4.8v

You mean C31, not ,c32 on the MPU?

Just as zaza said, 4.8v at TP2 on the driver board is too low.

Can you reseat J101 and J114 and remeasure the voltages?

J101 has burnt pins?

#19 3 years ago
Quoted from PinballManiac40:

You mean C31, not ,c32 on the MPU?
Just as Zaza said, 4.8v at TP2 on the driver board is too low.
Can you reseat J101 and J114 and remeasure the voltages?
J101 has burnt pins?

101 and 114 reseated.

Still at 4.8v

No sign of burnt pins at 101

IMG_20201227_150049 (resized).jpgIMG_20201227_150049 (resized).jpg

Battery holder off now

Some corrosion down there, I had treated it with corrosion gone (boric acid)

IMG_20201227_153510 (resized).jpgIMG_20201227_153510 (resized).jpg

I've had another go, rinsed and giving it a light touch off some Scotchbrite.

I wasn't too concerned about this, as it seemed light, had treated it and everything was working, up until the nvram went in. I'm baffled why 4.8v would allow it to work before, but not now, as I havent been over on the power board.

I guess, as you say, the very act of banging around with the boards and connectors and heating things up can bring out something that was hanging on for dear life

I have resistors, so I can replace those showing signs of corrosion, I'll give them a test.

I also didn't know BOP had time based Easter eggs, I'd been reading about Midnight Madness and the like, but hadnt seen BOP mentioned. I'll do some more reading.

But, 4.8v isn't enough juice so I need to fix that anyway?
IMG_20201227_153220 (resized).jpgIMG_20201227_153220 (resized).jpg

#20 3 years ago
Quoted from Biglouie:

No sign of burnt pins at 101

You need to look at the pins in the wired connector that goes into J101.

Quoted from Biglouie:

But, 4.8v isn't enough juice so I need to fix that anyway?

Too low and must be fixed. Anything under 4.9v will either cause resets or not boot up at all depending on how low the voltage is.

#21 3 years ago
Quoted from Biglouie:

I also didn't know BOP had time based Easter eggs, I'd been reading about Midnight Madness and the like, but hadnt seen BOP mentioned.

Never heard of anything before tied to the clock in BoP. Even if there was, the time will be random and still will function, but will just be off depending on how long the game is left on/off.

#22 3 years ago

All the parts in red that were hit with the battery corrosion will need to be replaced. It already entered the parts and no way to clean inside the parts. You can even see the resistors still are corroded. Marked in yellow are dark areas where the corrosion has been eating at the ground plane under the solder mask which is why it changed to a dark color. You have to sand it down to get to the source in order to get rid of it. Otherwise, it will continue to spread like cancer.
BoP (resized).jpgBoP (resized).jpg

#23 3 years ago

Here is one very similar condition board that I repaired with battery corrosion. I got lucky that it only damaged one IC and none of the resistors below the battery holder, but it got into the parts above the battery holder. I coated the board after, which is why it is shiny.

pasted_image (resized).pngpasted_image (resized).pngpasted_image (resized).pngpasted_image (resized).pngpasted_image (resized).pngpasted_image (resized).png

#24 3 years ago

Ooosch. That fills me with dread, but I'm up for a challenge!

I'll start working through some of the Pinwiki 5v reset stuff, see if it helps me get to the low 5v issue.

Thank you so much for sharing your expertise, I'm still stumped why it booted before the nvram swap, but seems there are a few elements to track down and fix up.

#25 3 years ago
Quoted from Biglouie:

Thank you so much for sharing your expertise, I'm still stumped why it booted before the nvram swap, but seems there are a few elements to track down and fix up.

Quoted from PinballManiac40:

Anything under 4.9v will either cause resets or not boot up at all depending on how low the voltage is.

What you can do is take this CPU board and test it in Hurricane or Getaway to see if it boots. Just disconnect the short ribbon cable between the CPU and driver boards, and only connect the J210 power connector on your BoP CPU board in either of the two games. Hold the board on the outer edges of the board and turn on the game. This will prove whether your NVRAM has a good install and if it is that the 5volts is too low to boot up your BoP CPU board in BoP.

#26 3 years ago

Board into Hurricane.

D19 and D21 same, both stuck on.

So I guess my nvram install is a bust! No surprises here I guess! Haha

I'll clean up the board and resolder the socket, there will be more fresh solder floating around, perhaps it will improve things.

Thanks again, appreciate the guidance to get me to this point.

Rob

#27 3 years ago

Thought I'd check BR2 following pinwiki 5v line of enquiry..

BR2, top left leg, gives 11.1v DC and 0.135v AC

Pinwiki says 9v DC and 300mv AC. Is the AC reading what people refer to as ripple? So low is good?

I'm out on both counts but not sure of that's a bad thing or not!

IMG_20201227_185215 (resized).jpgIMG_20201227_185215 (resized).jpgIMG_20201227_185221 (resized).jpgIMG_20201227_185221 (resized).jpgIMG_20201227_185556 (resized).jpgIMG_20201227_185556 (resized).jpg

J101 looks pretty clear of burns too, both plug and header.
IMG_20201227_185841 (resized).jpgIMG_20201227_185841 (resized).jpgIMG_20201227_185848 (resized).jpgIMG_20201227_185848 (resized).jpg

#28 3 years ago

You are not looking at the right side of J101. The shiny part of the pin should show thru the housing on the side you dont show in the pictures.

So you checked the NVRAM to the ASIC and to the game ROM as well? Make sure you test every pin also to the adjacent pin in the IC socket that you installed. Pin 1 to 2, pin 2 to 3, etc to be sure there is no solder short under the socket.

#29 3 years ago
Quoted from PinballManiac40:

You are not looking at the right side of J101. The shiny part of the pin should show thru the housing on the side you dont show in the pictures.
So you checked the NVRAM to the ASIC and to the game ROM as well? Make sure you test every pin also to the adjacent pin in the IC socket that you installed. Pin 1 to 2, pin 2 to 3, etc to be sure there is no solder short under the socket.

I made a list of 28 pins that are on the nvram, looked up on the schematic where they all go to (3 to ASIC iirc and most to cpu then w6, w3, gnd and r12)

IMG_20201227_204248 (resized).jpgIMG_20201227_204248 (resized).jpg

I've buzzed them all, top and bottom, all check out.

So, I'm thinking it must be something that has been damaged in this overall process.

I've buzzed the adjacent pins on nvram socket as you suggest as well, all good.

I think I need the number for Truckmaster, I'm not going to make it as a pinball technician..

IMG_20201227_205055 (resized).jpgIMG_20201227_205055 (resized).jpg

I'll change that for a Trifurcon tomorrow, see if it makes a difference to the 5v?

What decides the 5v ? It comes in on j101 then the board regulates it somehow? Do I need to test j101 to see what power is coming in perhaps? Be good to get up to 5v and learn how all that works!

#30 3 years ago
Quoted from Tallon:

so.....It worked before you worked on it and now it wont boot. always check your work. Make sure you didn't get a solder bridge under the socket.

THIS. If it worked before your work on U8 and now it doesn't work after your work on U8 and that was the ONLY thing you did then that is cause.

Quoted from Biglouie:

I made a list of 28 pins that are on the nvram, looked up on the schematic where they all go to (3 to ASIC iirc and most to cpu then w6, w3, gnd and r12)

I've buzzed them all, top and bottom, all check out.

I've buzzed the adjacent pins on nvram socket as you suggest as well, all good.

Buzzing top to bottom and side to side (adjacent) is not enough. The 28 pads on U8 do not connect to a single point. Without branching from a trace there are FOUR potential branches from a pin/pad. Up from the pad on the front of the board, down from the pad on the front of the board, up from the pad on the back of the board and down from pad on the back of the board. That also requires that the through hole connects the pad on the front of the board to the pad on the back of the board.

If you buzz for continuity on the pad from the front to the back that does not take into account the trace leading away from the pad. It only tests the continuity of the through hole.

Most of the connections on U8 are part of a bus so if you really want to check for continuity of the bus test the points that do *NOT* include U8. That way you know that continuity through the pads of U8 are correct. If there is a break noticed then you can test the pad on U8 to isolate if the through is damaged.

I would be willing to be you have a broken trace coming off a pad - either on the front or the back of the board (more likely the front as it is concealed by the socket you put in place). Testing for continuity will reveal the break. You can fix the break using a jumper or if you want to see the break you will have to remove the socket. Removing the socket will risk further damage. You aren't the first person who has done this (I am guilty of having done this several times) and you won't be the last. There's another System 11 thread where someone did the same thing. Changed U25 (SRAM) and now doesn't have a boot. It's the same thing. Broken trace. It ALWAYS is. The odds of having a bad "new" part is low. Very low. The odds of damage to the PCB is high. Very high.

If you want a visual guide I have attached some images that might help. Good luck. Finding it is tedious but finding it is also rewarding.

01_U08_data_bus.jpg01_U08_data_bus.jpg02_U08_address_bus0.jpg02_U08_address_bus0.jpg03_U08_address_bus1.jpg03_U08_address_bus1.jpg04_U08_local.jpg04_U08_local.jpg
#31 3 years ago
Quoted from Biglouie:

I'll change that for a Trifurcon tomorrow, see if it makes a difference to the 5v?

What decides the 5v ?

Comes in on the 2 red wires on J101. Hold for now. We need to be sure the full AC voltage is getting to the bridge rectifier. If the voltage is good at the bridge rectifier, then no need to change J101.

#32 3 years ago

Wow DumbAss I guess there is far far more to this than I could have ever imagined! Thank you for explaining that.

I'll have a look into the diagrams. I had expected to have to remove the socket.

The change was the nvram, I was so incredibly careful, that will be where the issue lies.

But the 5v also needs sorting.

#33 3 years ago
Quoted from PinballManiac40:

Comes in on the 2 red wires on J101. Hold for now. We need to be sure the full AC voltage is getting to the bridge rectifier. If the voltage is good at the bridge rectifier, then no need to change J101.

Can I test on the plug itself for AC or once it's plugged in the board on its way to the fuses?

IIMG_20201227_231421 (resized).jpgIMG_20201227_231421 (resized).jpg

I double checked the jumper Molex on the transformer just to check that (Getaway was still wor d for Europe when I got it), this one checks out for 240v

IMG_20201227_211839 (resized).jpgIMG_20201227_211839 (resized).jpg

#34 3 years ago
Quoted from Biglouie:

Can I test on the plug itself for AC or once it's plugged in the board on its way to the fuses?

Needs to be checked way beyond J101 connector.

#35 3 years ago
Quoted from Biglouie:

The change was the nvram, I was so incredibly careful, that will be where the issue lies.

This is why installing NVRAM (by removing the SRAM and installing a socket) is rated as HARD for WPC. I have had friends deliver boards to me to repair because they gave up figuring out what they did wrong (i.e. which traces, pads or through holes they damaged or removed). I don't bother figuring out what they did wrong either. I just remove the socket they installed and inspect everything.

I realize that learning how to desolder has to start somewhere but I would never recommend anyone learn on this board. The exception to this is if an experienced person is supervising or monitoring the progress.

If salvaging the original SRAM is not desired then an alternative that I do recommend is cutting the legs of the IC (as close to the body as possible) and removing the body. Proceed to removing each individual leg by heating the pad and using needle nose pilers to remove each leg. This removes most every possibility of damage to the pads and through holes. Of course ... that recommendation is too late for this scenario.

#36 3 years ago
Quoted from DumbAss:

This is why installing NVRAM (by removing the SRAM and installing a socket) is rated as HARD for WPC. I have had friends deliver boards to me to repair because they gave up figuring out what they did wrong (i.e. which traces, pads or through holes they damaged or removed). I don't bother figuring out what they did wrong either. I just remove the socket they installed and inspect everything.
I realize that learning how to desolder has to start somewhere but I would never recommend anyone learn on this board. The exception to this is if an experienced person is supervising or monitoring the progress.
If salvaging the original SRAM is not desired then an alternative that I do recommend is cutting the legs of the IC (as close to the body as possible) and removing the body. Proceed to removing each individual leg by heating the pad and using needle nose pilers to remove each leg. This removes most every possibility of damage to the pads and through holes. Of course ... that recommendation is too late for this scenario.

Definitely a harder job, the leg removal was the process I followed as well, but I thought applying fresh solder and the vacuum solder sucker would be ok. I've done other jobs, I think this one was just too delicate for my skill level!

#37 3 years ago

Oooooo, never a solder sucker on the WPC boards. I actually have a second desolder station with a micro nozzle just for WPC and newer work. I found an Ungar/Weller solder/de-solder station pretty cheap back in the day that I’m keeping limping along. It’s a version of the old hot vac system and does great on the tiny pads and leads for WPC stuff. It’s personal preference. I know some people who tried my setup and scorched the crap out of thru holes. Just gotta keep at it til you figure out what works for you.

#38 3 years ago

It's a Hakko 301.. I'll keep practising.. I have a board that is so badly battery damaged that it could never live again and a load of old arcade pcbs that are dead.. So I'm learning on those.

I'll be buying a pinled repro if this sanding turns into a pita. I enjoy the learning but it comes to a point where it's just too laborious!

I just want to learn how to do this stuff but I also want to play pinball. Haha

#39 3 years ago

For fun. I desoldered the socket to see what lay beneath.
IMG_20201228_172745 (resized).jpgIMG_20201228_172745 (resized).jpgIMG_20201228_172755 (resized).jpgIMG_20201228_172755 (resized).jpg

Pin 10 I had to jumper to the CPU, as it did test with no continuity when I was doing this. I can see why now

I've sanded it back, but can this be repaired as the pads gone and it goes out two ways.

IMG_20201228_173711 (resized).jpgIMG_20201228_173711 (resized).jpg

#40 3 years ago
Quoted from Biglouie:

can this be repaired as the pads gone and it goes out two ways

fixable...yes,
but you might go for the easy way and put a jumper in place.
If I count correct, it is address A0, going to U9 pin 25 ---U8 pin 10 --- U6 pin 12 (please verify)

address A0 (resized).jpgaddress A0 (resized).jpg

#41 3 years ago

Buzz test confirms both correct for U8 pin10 by dropping on to the tiny amount of trace on front side.

Assuming I can get the socket back on again without causing further damage, I'll give it a go and jumper the back!

Thank you zaza

#42 3 years ago

Kynar wire works wonders for small trace repair.

#43 3 years ago

Ok, it's a bit of a flux fest but stage 1 complete.

I'll test and do the jumper..

IMG_20201228_191807 (resized).jpgIMG_20201228_191807 (resized).jpg

#44 3 years ago

that graphic from dumbass was EPIC!

#45 3 years ago

I think A0 also goes to u4 Pin8. But that buzzes.. so all good.

Feel like I'm getting closer.

IMG_20201228_193942 (resized).jpgIMG_20201228_193942 (resized).jpg

#46 3 years ago

Correct, but there is no reason to add a jumper to U4-8.

As soon as the jumper between U6, U8 & U9 is installed, you will find continuity between all ICs address A0 (U1, U4, U6, U8, U9)

#47 3 years ago
Quoted from Biglouie:

Ok, it's a bit of a flux fest but stage 1 complete.
I'll test and do the jumper..
[quoted image]

I use chip-quick to lower the temperatures. Its saved me a lot of headaches.

#48 3 years ago

No one is more surprised than I. Thank you @zaza, giving me A0 to ASIC was the missing piece of the puzzle.

IMG_20201228_200224 (resized).jpgIMG_20201228_200224 (resized).jpg

#49 3 years ago
Quoted from pinballinreno:

I use chip-quick to lower the temperatures. Its saved me a lot of headaches.

Good tip. I've got some in the workshop .. I was planning to use it if I ever needed to change C5 on a power board.

After the last 2 days.. I'll be staying away from that for a while!

#50 3 years ago

: ) Congrats !

Promoted items from the Pinside Marketplace
From: $ 1.00
Playfield - Other
Rocket City Pinball
Other
There are 51 posts in this topic. You are on page 1 of 2.

Reply

Wanna join the discussion? Please sign in to reply to this topic.

Hey there! Welcome to Pinside!

Donate to Pinside

Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


This page was printed from https://pinside.com/pinball/forum/topic/wpc-stitching-a-through-hole-advice-sort?hl=biglouie and we tried optimising it for printing. Some page elements may have been deliberately hidden.

Scan the QR code on the left to jump to the URL this document was printed from.