Taito Ice Cold Beer Total Restoration, PCB Repair and Upgraded Sound & Selectable Free Play Mode

jimmydeanbean

Member
Joined
Apr 25, 2022
Messages
57
Reaction score
35
Location
Wisconsin
Taito Ice Cold Beer Inoperable, repair thread, but backlogged. Roughly a year of backlog…

So I'm making an Ice Cold Beer repair thread. I acquired one for myself, my brother also acquired one around the same time. My brothers functions but has a couple bugs during gameplay, it will randomly freeze and needs to be reset. Being a good brother I agreed to figure out the issue with his game as well.

I have at this point electrically repaired both machines and have fully cosmetically restored one of them. I also made some upgrades to the PCB to include a switchable free-play mode and normal game play mode, via a hacked EEPROM chip with a custom PCB daughter board.

I'm working on the 2nd restoration. The 2nd will be faster and easier as it was not in as terrible of shape physically, even though it looks like someone may have hit it with a baseball bat. So the following will be a chronicle of that work. First electrical restoration, then cosmetic. I will unfortunately be missing many pictures I took of the electrical restoration as my phone, which had all of the photos on it, was stolen. So I apologize for any lack of electrical photos but most of this will be evident from the schematics in the owners manual.

As I began my journey down this rabbit hole, maybe rabbit hole is not accurate, black hole may be more accurate. Anyways, I found the below thread incredibly helpful to better understand this machine.

It chronicles the journey from completely inoperable to fully functional of an ICB machine. I learned quite a bit from it but my issues were spread across two different machines.

The following posts will be a backlog of posts that I meant to make but didn't, over the last year. The goal here is so you can gain some Dos, Don'ts and some confidence to go from zero experience in this field to doing board repairs, cosmetic repairs and the such. This post will be a work in progress that I hope to write fully over the course of the next month or so. So I'm going to keep at it until it's done.
 
Backlog begins now:

So 1st part of backlog is regarding the power supply on this machine.
When I received this machine in, my first inkling was to power it up and see what I've got. I was told when I purchased it that the machine would power on but wouldn't coin up. I thought maybe it was just a bad switch, or bad chip that handles that input, a quick fix! Ehh… Yeah it wouldn't coin up but other than make a series of strange noises I wouldn't classify it as powering up.

The machine had been retrofitted with one of these new modern power supply's you can find on Ebay. I've read a few other posts about it causing more problems than it fixed. Maybe it works fine but I want this machine to be as original as possible. So I pulled the thing out and I now focused on the original power supply.

I started by checking the fuses on the old power supply and knew I was in for a rough ride. The mains fuse is supposed to be a 3A fuse and there was a 10A fuse in it, it was blown… Also one of the 6A fuses and one 4A fuse was blown.

I began by separating the supply from the main PCB and the rest of the machine. I wanted to figure out what was causing the fuse to blow. I replaced the fuses with the proper sized fuses and plugged it back in. Instantly blew one of the 6A fuses. These fuses run through the two large electrolytic capacitors. So I figured either it was one of these capacitors or a diode bridge that it feeds.
How do you diagnose which is which? After spending a few hours online I found this very helpful video on how to diagnose diode bridges with a simple voltmeter.
(444) How to test a bridge rectifier with a multi-meter - YouTube

The last part is great as it shows exactly what values you should get between which pins. Prior to learning how to do this, I bought ahead of time a few extra diode bridges figuring one would be bad. They weren't, all functioning as intended. Some of the diode bridges on this supply are not listed in the manual. In this blog I will attach at some point a true parts list for all electronic components. Stay tuned for this.
 
So what the heck is wrong with my power supply?

On to the electrolytic capacitors next. These things are old. The big one is actually in a ceramic case! Pretty cool stuff. So, how do I test these to figure out what is good and what is not?
There are as many ways to check a capacitor as there are flavors at Baskin Robbins. But after boiling down blog after blog on the subject, the easiest was with an ESR (Equivalent Series Resistance) meter. This allows you to check a capacitor while it's still in the circuit. I figured this would be helpful later for checking the soldered on-board capacitors. I went with the, Signstek MESR-100 V2, on amazon. It seemed pretty decent and it wasn't super expensive.

A quick note here. There are 2 large capacitors and one small capacitor on the main power supply. The small 1000uF capacitor is an add on from a service bulletin. This was added to ensure the errant ball solenoid had enough oomph to get the ball up to the playfield. If you don't have this on your ICB you should add it. The one on mine is a 1000uF 63V cap. The service bulletin calls for 1000uF 60V, attached is the original service bulletin so you can read it in it's entirety.
Now the two large capacitors are called out in the original guide as a 50,000uF Cap and a 20,000uf Cap. What voltages are they rated for though and what is actually on the board? We'll I'll tell you it doesn't match and the guide doesn't list the voltages! A lot of things don't exactly match the schematics on this game so get used to it.

What is actually on the board, 1 – 51,000uF 25VDC Cap and 1 – 18,000uF 30VDC Cap.

I checked all three caps with my fancy new ESR meter, I compare the values I read with an ESR table. ESR tables give you a rough idea of the ESR value you should get with a given capacitance and voltage capacitor. Everything looked good. Okay…. Now what?
 

Attachments

  • 20220504_174804.jpg
    20220504_174804.jpg
    168.5 KB · Views: 14
I looked at which fuse blew and traced it to the large 51,000uF cap. I went on to the next way to test a capacitor, which is outside of the circuit. I removed it and used a multimeter in capacitance mode. The battery from the multimeter should slowly charge the capacitor and raise the resistance of it. I got nothing from it, dead short! Okay! I'm onto something now.

After checking, checking and checking some more you cannot find a 51,000uF 25 VDC capacitor. Okay so a quick note here as well, these large old caps have a WIDE range, -10% - +75%. So don't be too picky about a replacement capacitor for this system as modern caps will have a much tighter tolerance at this size. What I learned after much asking around though is the General rule of thumb is to keep the capacitance the same but if needed you can go up on the voltage. I was able to find a brand new 51,000uF 40VDC capacitor. I installed the capacitor, put a new fuse in and eureka the power supply lives. I had voltage at all of the test points, no blown fuses and now I'm on to the next piece, the main PCB.

A note about capacitors here as well. I was told that the probable cause for this caps failure was due to it probably sitting unused for many years and then being used. Capacitors like this one are essentially aluminum foil that has been "rusted" in a controlled fashion to grow an aluminum oxide layer to a certain thickness. That provides the dielectric separator to create capacitance. When the aluminum is not cycled for a long period of time, that oxide layer can degrade and eventually cause a dead short. If you want to do this properly look into capacitor reforming. It will preserve those old caps if they haven't been cycled in years. Great reading and a good piece of knowledge to put in the ol' bag o tricks.
 
Now onto my PCB shenanigans. I started with my machine, the non functioning one. Which thankfully made it much clearer on how to fix the 2nd machine later on.

For reference I have only ever built a PCB from a kit before. I have soldered IC chips and resistors to a PCB, but I have never de-soldered anything other than a resistor in my life.
So I made some purchases and in my mind investments to make sure I can keep this game running in the future and who knows maybe I'll be crazy enough to dive into another game at some point.
I won't go into all of the research but I decided to get three things:
Hot air rework station
Good soldering iron
Portable Desoldering Tool

So here is a moment where I learned a lot, I didn't buy everything at once. I started with a soldering iron and hot air rework station. My first attempt to de-solder didn't go well which I'll chronicle later but that led me to the portable desoldering tool. It is a soldering iron with a built in pump. It works great! I came up with a method for desoldering which I'll describe in detail later, it works, use any method that works for you but this one works well for me and I haven't hurt the board yet.
 
So I plug my main PCB back to the wire harness and my newly rebuilt power supply and turn on the machine. I hear a humming noise followed by a thumping "heartbeat" every 700ish ms, and a few lights are blinking. The main display shows a single zero in each line of the play field.

I put the machine into the diagnostic mode by grounding out Test Point 17. To my absolute astonishment, several components checked out! The Display test's are good, the RAM exercise is good, both Eproms are good, all switches worked fine. After that the I needed a scope which I didn't have at the time to check further. I moved onto the sound test, it made a horrible terribly loud noise. After that I turned it off. Okay so the main parts of this thing are good, now what?

I wonder if there is something shorting on a lamp or switch. I disconnect some of the wire runs that connect the switches and lamps, power back on, and same thing. Hmm..

Here is where I make a WTF are you doing mistake, It was late, I was tired and should have stopped an hour earlier in my debugging. I plugged those wire harnesses back in while the game still had power on it. Not only was that just a bad idea, I also had the connector turned 180 degrees. It wouldn't fit together but one pin was able to touch just enough and boom big spark. I'm done for the night. I just made my board problem even worse.

Now when I power it on, the display has jumbled letters flickering, no lamps are flickering and everything seems worse.

What did I just do?

A few days go by and I work up enough nerve to work on my poor ICB machine. I knew I messed it up but how bad?
 
I put it back into test mode. The Display test still good, RAM good, Eproms good. I went to the switches next but I didn't cycle them, I just checked each phase of the test to see if anything was stuck closed or if there was anything else really weird looking. Knowing from the schematic that the lamps are controlled by the Sound chips I/O I figured since that is what probably blew sparks, I would focus here next.

I wasn't sure where to begin since at this point I don't have a Oscilloscope to take a deeper dive into this board. I figured I'd start with something basic, something I learned from the other ICB repair thread I read. Some of the chips on that board were shorting direct to ground on pins that shouldn't be going to ground. Since I also had a perfectly functional board on my brothers unit I figured I could quickly test each pin on a chip on my board, then compare that to my brothers board and note any differences. Doing so I found 4 chips that had pins going to ground that the "good" board did not. Chip U3, U4, U10 & U26. U10 & U26 have some pins that connect to the DIP switches, so there could be differences there depending on settings.

I got my hands on an oscilloscope and went through all of the test listed in the manual. I didn't make notes on this part of the process unfortunately but mostly it didn't show much. The main parts of the machine were functioning which is what the oscilloscope is there to verify. The only thing I found was a noisy test point, but this condition seems to be normal across both machines even now when fully functioning.
 

Attachments

  • 20220529_141613.jpg
    20220529_141613.jpg
    190.4 KB · Views: 3
I later found that chip U26 was just fine, there was some DIP switch settings here that were different. So now were down to these three chips.

U3 & U4 were definitely fried, either originally or from my late night screw up. Also U10 was fried as well. Which all makes sense once you start to understand the schematics of this game. U10 is primarily part of the sound system. There are two chips that handle the sound U9 & U10. They somehow together phase in sounds that are summed together to create the game music and noises, it's an interesting system. These chips also have I/O capability, which handles most of the lamps and input switches. Behind these chips are the PIA (Peripheral Interface Adapters), and behind that the main CPU. I'm hoping my issues end with the 3 chips I've identified but who knows it could go all the way up to the main CPU.

Knowing that I could be entering the great unknown with this board, at this point I decided to make a list of all components on this board, as well as current replacement parts. This is pretty comprehensive and includes all sources that I was able to find for replacement parts. The hardest to source is the sound I/O. These chips have not been produced in a long time but there have been some different remakes of the chip, in a production setting and as cheap knock-off's. So use caution when trying to source these things. In the list I have here it does show where I sourced mine from. They work perfectly and the same as the original chips.

I have some good information on the power supply PCB as well as some of the differences between the schematic and actual builds on these boards.
 

Attachments

  • Component List.zip
    38.9 KB · Views: 19
Now back to my troubleshooting. So how do we know what to fix here? At this point it's a bit of trial and error.. I figure let's start with the chips we know are bad. I tried U3 & U4 to begin with.

Now, I have never replaced an IC chip on a board before. This board is especially not that great to begin with. It's rather large and flexible, which makes it susceptible to breaking traces or solder joints just by moving it around. It's not a lot of fun to work with. But Here we go anyways. I began by using the hot air rework station. I began putting some flux all over the IC chip I needed to desolder. After heating the thing slowly I couldn't get the thing to come loose. I would get the solder in its liquid state on most of the pins but not all of them simultaneously. After a while of burning flux and the board itself I stopped before I did further damage. I thought about just getting a tiny pair of side cutters to cut it out physically and then desolder the pins but I couldn't find anything that looked like it would work.

Now que my saving grace, the portable desoldering tool. This is a really amazing little piece of equipment. It's a solder gun that has a suction piece to the gun itself. You use the gun to individually heat up a solder joint on the chip, then once the solder is liquified you hit a button that turns on a pump to literally suck the solder out of the joint. This will remove 95% of the solder. What I also discovered here and why it was impossible to remove the chip with the hot air rework tool is the pins are actually bent over on the board. I'm sure this was a way to keep the chip in place during soldering and cooling of the solder. Anyways now that you can treat each pin individually you can also use the hole in the gun to grab the pin itself and straighten it out, (while the remaining solder is still liquid). You do this with each individual pin until all of them are straight and mostly free of solder. You'll likely have a pin or two with just a little solder left around the pin, you'll need to look closely at each pin until you free them completely. DO NOT try and force a pin with tools or otherwise free from any remaining solder, it can and will pull the trace from the board itself. Use the desoldering tool to reheat the pin and suck out any remaining solder. Now you wont likely still be able to get that chip out as no matter how much you try, there will still be enough solder around enough pins that it's not 100% free. Again DO NOT try to pull them out, you'll damage your board.

Here is the specific tool I bought.

Now that the pins are straight and mostly free of solder, you can go back to that hot air rework station. Glob up that that IC chip with some flux and get it heated up, now we can free that IC chip with very little time and effort. The traces are all intact and the board is ready to accept a new chip.

Now I did make an upgrade to this board. Any IC chips that I removed I made the decision to put a DIP socket in place of the chip I had to replace. This soldered in DIP socket allows me to replace a future chip with no desoldering required. Hopefully I won't need to take advantage of that, but as I plan on keeping this machine for the rest of my days, it's quite possible something will fail on me in the future. In fact from the blogs I've read on this machine, most issues seem to stem from the sound chips and their downstream transistor array chips. So it's quite likely that will be something that fails in the future.

Doing this also allows you to put the heat into the DIP socket as opposed to heating the chip by soldering each pin of the chip. This puts less wear on the chip itself from day one.

Anyways, I have now replaced the two chips U3 & U4, and put in the new DIP sockets. I turned on the machine to see if anything has now improved. I figure it's possible that the upstream problem with sound chip could short out my new U3 & U4 chips, but I bought a bunch of them and I now have DIP sockets so if I have to swap them out again it's not a big deal.

Success, I'm now back to where I started, I hear a humming noise followed by a thumping "heartbeat" every 700ish ms, and a few lights are blinking.
 

Attachments

  • DIP Socket.png
    DIP Socket.png
    31 KB · Views: 1
Last edited:
Next up time to replace the sound chip U10. I was able to source some unused old stock of these sound chips, they are not super common these days. So with my new old chips in hand I replaced U10 chip and plugged it back in to see what happens.

**Now here is where I didn't realize I had actually fixed the board and it was perfectly functional but I thought it was broken still.

It turned on, the heartbeat sound was gone and wasn't doing much. I thought maybe it was fixed. So I hit the "free play" button the previous owner added to the machine, more on this later but it was an ugly button drilled through the coin door, I hate how it looked. Anyways, I hit that and there's this horrible static noise playing on the speaker. I thought the switch shorted it out or something. It was a loud terrible noise.

I unplug the machine and start to think what else is wrong with this thing.

I cut out and threw into the trash where it belonged the ugly button added as a free-play button. And then checked all of the wiring attached to the coin up circuit. Everything checked out. Nothing was grounding out, and nothing seemed to be wrong. I plug in the machine and same thing, I get this horrible static noise when I hit the coin up switch on the coin door.

I thought maybe the other sound chip U9 had some issues as well, it's the other half of the sound so I swapped it out with a new chip and powered it back on. Same thing but here I now realized, after hitting the coin up twice I got a credit. The soft pulsing glow of the start button beckoned me to it. I hit start and away it went the machine started playing 100 bottles of beer on the wall and went through the start procedure! My machine was working, electrically speaking! But I still had that darn static noise, anytime I added a coin, anytime I lost a game and if I made it all the way through all 10 holes.

I made some posts on a separate thread asking if anyone knew if this noise was caused by anything in particular. Ice Cold Beer Static Noise on Counters | Museum of the Game® Forums (arcade-museum.com)
You can see this here.

Final update, this is an intentional noise, it's supposed to sound like someone opening a bottle of beer! Part of the problem with ICB in general is the volume control on this machine is crap. If you adjust the volume control on the board, It's either off or a cat hair past off, it's louder than a Metallica concert. Since it's always too loud most people keep that volume way down in order to hear yourself think, as a result most people don't notice this noise as it happens almost exactly when a coin falls into the coin bucket. Weather your coin bucket is full or empty it makes a fairly loud enough noise to drown out the static noise that represents that beer bottle opening. So I've now gotten a fully functional ICB board and power supply. The physical game itself needs some work but I feel like I could take on the world now. I, without any real prior technical experience in board repair, diagnosed and repaired a board and power supply from 1983. Heck yeah!
 
Now I decided at this point, now that I know my board is good and that it makes a horrible noise when I put a coin into it, I want to fix this problem with the game. I like the noises made during gameplay but this static noise is terrible and I cannot get the volume right. The potentiometer on the board has some serious issues with it. It has some dead spots and is making some noise itself. Now I found some drop in replacement potentionmeters that are made specifically for this board. However, I decided I didn't want to use that, if I'm going to put the time into desoldering this thing, which was WAAYY harder than I anticipated, I want to put a better POT in there. After a lot of research it seemed this a CERMET potentiometer was the best bet for a very reliable device. It's also sealed so dust will not make its way onto the wiper, it's as bullet proof as you can get. Below is a link to the one I sourced. It was CERMET and had leads long enough I could get them to the terminals on the board.

14910A0BHSX10502KA Vishay / Spectrol | Mouser

Now after trying to desolder the old potentiometer I will say I learned a much better way to get the thing out. Don't try to desolder it. Just like the chips the four large pins that attach this to the board are bent over, so you won't be able to heat it up. I tried my handy portable desoldering tool and I couldn't suck out enough solder to get the pins free. They were also large and not easy to unbend. It took me a while but I heated each pin up enough and then with plyers unbent the pins. I globbed it up with flux and tried to desolder it. The end result was I pulled one of the solder joints out of the board entirely. It came out but it wasn't pretty and I had to make some minor repairs to the board to get the traces to work properly again. IN the future if I had to do this again. I would surround the board with some cloth or light towels and use a small Dremel tool to literally cut the pins of that POT and then desolder each pin individually. It would have been messy but it would work and not heat damage the board.

Now, I used the wiring diagram the POT came with and matched that up to the schematic of the board itself. Now that I have my new CERMET POT installed I powered up the machine. This did improve the condition of zero volume and too much volume. There was now no dead spots on the new POT so I could keep it just above absolute zero and get some volume, a touch above that and it was too loud again. I wanted to fix this issue that was from the manufacturer on day one.
 
I looked at the amplifier for the sound itself to see what could be done here. Maybe there was a way to modify it so the sound was normal. No such luck. I found the original schematics for the TDA2002H, audio amplifier. This manufacturer specs match exactly how this was setup on the board of the ICB machine. So the only thing that I did note was that the speaker used on the machine was a 4 Ohm speaker. Thinking outside of the box a little bit I figured maybe we could go with a higher Ohm speaker, essentially cutting the amount of current at any given volume level in half. This should give me greater control over the volume. So I found an 8 Ohm speaker that had the same hole pattern as the current 4 ohm version. Once I pulled the old speaker out it was a good thing I did, the paper cone had almost completely disintegrated. I'm amazed that the thing actually still made any noise at all.
FR 6,5" - 8 Ohm Visaton | Mouser

Now I have my new 8 Ohm speaker soldered into the original connector. I power the unit up and Eureka, I now have a much better control on the volume. It's still touchy on the POT as it doesn't require much movement with the POT to get a noticeable change in volume but I'm now able to get the volume to manageable level. And that loud static noise? Now it sounds kind of like a beer bottle opening. It's volume is in-line with the rest of the game noises. I finally have 1 working and upgraded PCB and power supply. I actually have another PCB upgrade I did much later in the process but I'll go over that later.
 
Now onto machine #2 and it's issues. This seemed to me at first to be a daunting task. The game worked just fine, until it didn't. I couldn't predict when it would fail. When we originally noticed this problem, before I began reworking these machines, it was during a party and it was sporadic. Sometimes it was a few minutes into the game, other times it was after someone got the ball to hole 10. But it would eventually fail.

When I started I got lucky and the machine started failing after about 5 minutes of gameplay. Then a couple of minutes, then less. The problem was getting worse. I noted that it would just freeze up and some of lights on the playfield would light up. Not all of them, just some of them. This was similar to what happened to my first machine except right at power up. Which led me to believe it was one of those sound I/O chips again. I theorized that the sound chip would heat up during gameplay and eventually internally a circuit would open or close that wasn't supposed to and it would crash the game.

So I started by pulling U9 and swapping it out. I powered back up the machine and started to play. Ope, 5 minutes into gameplay and the thing stopped again. I pulled the board and now swapped out the U10 chip. Powered up the machine and played a nice long game! No issues, the game works perfectly.

That was easy! Looks like I might know a thing or two about this game now!!
 
Now that the game is working perfectly, electrically speaking anyways, I stumbled across a thread on ICB ROM files that are hacked to include free-play mode. Essentially when you power up the game always has the max credits in the machine so no one ever has to drop coins in this thing. I thought that would be a cool feature to have for a party, so I don't have to have a roll of quarters just sitting there. I also didn't have to add an ugly cyclops button to the coin door for a manual free-play. Original thread below:

https://forums.arcade-museum.com/threads/ice-cold-beer-freeplay-hack.146787/ Credit to DogP on the work done here.

I took this one step further though. The hack involves a variation of this tempest hack thread.


The original ROM file on chip U23, is on a 27c64 EEPROM chip. If you get a 27c128 EEPROM chip you can take both the original ROM file and the hacked version and combine them into a single file to use on that new chip. From there, there is a pin that you need to supply voltage or ground to. This activates the program in one half or the other of the memory of EEPROM chip, this creates a selectable program between original and hacked versions of the program. I created a little breakout board that the modified chip could sit on, install into the main board in place of the original chip and have a remote wireable switch to toggle between the two modes. Below are the parts I used for the breakout board. And in the above thread, the latest version of the EEPROM file that includes original and free-play mode.

How do you combine those two files you might be wondering, it's pretty easy and involves some command line code. I tried it though and it didn't work for some reason. So I sent it over to the guy at Hobbyroms.com and he combined the files for me. If you're interested in doing it yourself, read the following blog on it.


Below are the parts I used for the breakout board.


Here is the link to have the breakout boards made. It's a minimum run of three pieces from OSHpark.



Please note I have not attempted nor do I suggest you attempt to switch between the two game modes while the machine is powered on. In theory it shouldn't hurt anything but, I wouldn't risk it. I put the switch on the inside of the cabinet in a difficult spot to get to unless you have the key and can open the back panel.

I made a small mount out of a piece of metal I had laying around. The switch is mounted onto the piece of metal and attached to the inside of the cabinet. I ran a ground wire from the grounding lug on the playfield to the metal bracket I made, just-in-case you want any new electrical your running to have a fail safe ground connection.
 

Attachments

  • 372157590_698923832262385_4286701508730950541_n.jpg
    372157590_698923832262385_4286701508730950541_n.jpg
    136.2 KB · Views: 18
  • 373082745_840214071064317_1506260410886637313_n.jpg
    373082745_840214071064317_1506260410886637313_n.jpg
    270.3 KB · Views: 16
  • 373465972_710897084212354_5342734600256573189_n.jpg
    373465972_710897084212354_5342734600256573189_n.jpg
    215.1 KB · Views: 17
  • 373488003_675740317800307_6822837597096647369_n.jpg
    373488003_675740317800307_6822837597096647369_n.jpg
    170 KB · Views: 17
  • 1693884327331.png
    1693884327331.png
    162.7 KB · Views: 18
Last edited:
Back
Top Bottom