Repairing Martin Galway's Commodore 128

Summary

In this video, The 8-Bit Guy embarks on a fascinating journey to revive a unique Commodore 128D belonging to legendary chiptune musician Martin Galway. After Galway moved to Austin and handed over his non-functioning computer, the adventure begins with an exploration of its European features, challenges posed by the UK power system, and a series of troubleshooting steps. Enlisting help from June, a C128 expert, they tackle power supply issues, monitor compatibility hurdles, and finally, a tricky disk drive repair. This technical odyssey not only restores a piece of computing history but also highlights the complexities and rewards of vintage computer repair, culminating in Galway's joy at seeing his beloved machine return to life.

Takeaways

• 😀 The computer belongs to famous video game musician Martin Galway
• 💻 It's a European Commodore 128D that was brought from the UK to the USA
• 🚧 The computer wasn't working properly so Dave offered to try and fix it
• 🔌 The power supply was damaged during shipping to the USA
• ⚙️ With help, Dave diagnosed and replaced the broken power transformer
• 📀 The floppy disk drive was also found to be non-functional
• 💾 The disk drive mechanism was replaced with a working one
• 👍 After repairs, the computer and disk drive were operational again
• 🖥️ Despite trying many monitors, only an HDMI converter displayed properly
• 😄 Martin was happy to get his computer working again

Q & A

Who is Martin Galway and why is his computer significant?

Martin Galway is a famous chiptune musician known for his work on over 60 video game scores, particularly on platforms like the Commodore 64, NES, and Amiga. His computer is significant because it personally belonged to him and was used in the creation of his music.

What makes the European 128D computer mentioned in the script unique in the USA?

The European 128D is unique in the USA because it was never officially released there. The US received the 128D cost reduced version, which was made of metal and not considered portable, unlike the European version.

Why did Martin Galway decide against cleaning and retrobriting his computer?

Martin Galway decided to keep his computer in its original condition due to the various stickers on it, which likely held sentimental value or marked significant events or achievements related to his career.

What were the main challenges faced during the repair of Martin Galway's computer?

The main challenges included dealing with a computer that wouldn't power on correctly, a damaged 9V transformer likely caused by rough handling during shipping, and difficulties finding compatible PAL equipment in the USA for testing.

How was the problem with the 9V AC supply in the computer resolved?

The problem with the 9V AC supply was resolved by diagnosing a physically broken transformer and then replacing it with an external power brick, after an unsuccessful attempt to repair the original transformer.

Why was a modern video capture device used during the repair process?

A modern video capture device that supports both PAL and NTSC was used because PAL-compatible monitors are rare in the USA, and this device allowed for the testing of the computer's video output.

What discovery was made about the disk drive during the repair?

It was discovered that the disk drive, which had not been used for a long time, was not working. The drive mechanism was ultimately identified as the problem after testing with a known-good drive mechanism.

How did the team manage to find a replacement for the broken transformer?

Finding an exact match for the broken transformer proved to be impossible due to specific requirements. A close match was eventually found, but it required creative mounting inside the computer.

What was the outcome of the repair efforts on Martin Galway's computer?

The repair efforts were ultimately successful, restoring the computer to working condition, although the original monitor remained unrepaired due to the challenges of finding compatible PAL equipment in the USA.

What does the script reveal about the differences in video game development hardware and practices between the UK and USA?

The script highlights the challenges of working with hardware that was region-specific, such as PAL vs. NTSC video standards, and the unique practices of video game development, like using a separate development system connected via cable for code testing, which was a common practice for Martin Galway.

Outlines

00:00

🎮 Introducing a Special Computer

The video kicks off with an introduction to a unique computer belonging to Martin Galway, a renowned chiptune musician known for his contributions to video game music across various platforms, including the Commodore 64, NES, and Amiga, among others. The computer, a European 128D, holds a special place not just for its hardware but because of its association with Galway. Despite living in the UK, Galway has moved to Austin, leading to the computer's current condition of not working properly. The host, intrigued by the machine's history and Galway's legacy, prepares to examine and possibly repair the computer, while respecting Galway's wish to maintain its originality despite its yellowed appearance and unique power requirements due to its European origins.

05:03

🔍 Troubleshooting and Initial Fixes

Upon attempting to power on the computer for the first time, it's clear there are issues, prompting a disassembly and investigation into the machine's internals. The European 128D presents several challenges, including a different voltage requirement and a PAL video format, complicating the repair process in the US. Initial troubleshooting reveals a problem with the 9V AC supply, leading to innovative solutions involving a NuBrick power supply. Despite these efforts, the computer fails to start, indicating multiple faults. Assistance from June, an expert on the C128, uncovers a loose transformer as a significant issue, which is creatively resolved by adapting an external power brick, leading to a temporary fix that allows the computer to power on, albeit with new challenges presented by an incompatible monitor.

10:05

🛠️ Advanced Repairs and Solutions

The repair journey continues with the discovery of a non-functioning disk drive, an essential component for loading software. Diagnostic tests reveal the drive mechanism as the culprit, necessitating a replacement. However, finding a matching drive mechanism proves difficult due to compatibility and availability issues. The repair strategy includes adapting parts from a personal 1571 drive and extending wire connections to fit the unique setup of the 128D. This meticulous process highlights the challenges of repairing vintage technology, especially when dealing with rare or region-specific hardware. The successful repair of the drive mechanism marks a significant milestone in restoring the computer's functionality.

15:07

🎉 Final Touches and Conclusion

The final steps involve addressing the computer's display issues, a persistent challenge given the scarcity of PAL-compatible monitors in the US. After experimenting with various monitors and connections without success, a solution is found using a video signal converter, enabling a functional display. Martin Galway's return to retrieve the computer brings the project to a satisfying close, despite the unresolved monitor repair. The video concludes with an invitation to the audience to follow more repair adventures on The Time Rift Arcade YouTube channel, extending the journey of vintage tech restoration beyond this single computer.

Keywords

💡Commodore 128

The Commodore 128 is the vintage computer that the video focuses on repairing. It's an 8-bit home computer released by Commodore in 1985 as the successor to the popular Commodore 64. The script mentions the different variations like the 128D and European 128D models.

💡Martin Galway

Martin Galway is the famous chiptune musician who the Commodore 128 belongs to. He wrote music for many classic games in the 80s and 90s. His ownership and use of the computer provides historical significance.

💡Power supply

The computer's power supply transformer was found to be broken during diagnosis. Replacing and remounting this hardware component was crucial to getting the device functioning again.

💡Disk drive

The 1571 disk drive built into the Commodore 128 was not working properly. Identifying the faulty mechanical parts and replacing the drive mechanism was key to restoring data storage capabilities.

💡Diagnostics

Running diagnostics tests on components like the logic board and disk drive provided critical information to narrow down the issues during the troubleshooting process.

💡Retro computing

The practice of restoring and preserving vintage computers is an overarching theme. The video highlights various techniques used in the hobby to repair legacy hardware like the Commodore 128.

💡Chiptunes

The chiptune digital music, created by manipulating sound chips in computers like the Commodore 64, provides cultural context for Martin Galway's musicianship and computing history.

💡Vintage display

Getting an appropriate vintage PAL monitor and display compatibility presented ongoing challenges given the differences between European and North American standards.

💡Repair

The full process of diagnosing issues, testing components, replacing hardware, and restoring functionality is covered extensively regarding the Commodore 128.

💡Preservation

Preserving and elongating the life of aging computer equipment with historical ties is a driving motivator behind retro computing hobbies and repair efforts.

Highlights

Martin Galway is a famous chiptune musician known for his video game music on systems like the Commodore 64 and Amiga.

The computer belonged to Martin Galway and was not functioning properly, so he asked the narrator to take a look at it.

The computer is a European PAL version of the Commodore 128D, which was never released in the USA.

The power supply was found to be damaged, likely from rough airline luggage handling during its shipment to the USA.

With help from June, the faulty power transformer was replaced with a substitute to get the system functioning again.

The video output was scrambled due to PAL incompatibility, so a video converter device was required to properly display video.

The floppy drive was not working and hadn't been used since around 1989.

The floppy drive logic board tested fine, so the issue was isolated to the mechanical drive mechanism failing.

The faulty 1571 mechanism was replaced with a working one transplanted from the narrator's personal drive.

Wires between the replacement mechanism and logic board were lengthened by soldering wire extensions.

With the repaired power supply and replaced floppy drive, the system was restored to full working order.

No compatible PAL monitor could be located, so a video signal converter was used to allow display output.

Martin Galway came by to pick up the repaired system and seemed happy it was working again.

Repairing vintage electronics has overlap with repairing arcade machines covered on the narrator's other YouTube channel.

The narrator enjoys producing repair videos and aims to upload more to his arcade machine focused channel.

Transcripts

Hello, and welcome back to the 8-Bit Guy.

So, today we're going to be having a look at a very special computer, not so much the

computer itself but because of who it belongs to, so you've probably already read the title

of the video and you know this computer belongs to Martin Galway.

In case you don't know who Martin Galway is, he's a famous chiptune musician from back

in the day.

He has over 60 credits for famous video games, a lot of them on the Commodore 64, but also

on pretty much every sort of system out there ranging from the NES to the Amiga.

Later on in the 90s, he did music for MS-DOS games like Wing Commander II and Ultima VII.

here's a picture of Martin at work and if you look behind him, you can see part of the

computer that I have here.

So, now you're probably thinking, "but wait a minute, doesn't Martin Galway live in the

UK, hence the UK power connector on the back of this computer?"

Well, yes he did but a few years back he moved to Austin and I met up with him there a few

months ago and he asked me if I would take a look at this computer because it is not

working properly anymore.

So, let's have a look at it.

This is my first time actually handling a European 128D.

Remember here in the USA we never got this machine, we got the 128D cost reduced version

which was made of metal and was not considered portable in any way.

OK, here's the keyboard.

Yikes that is yellowed.

Since this thing runs on 240 volts, I'll need to use this transformer with it.

And since it is PAL, I'll also need to use this PAL monitor with it.

Martin also gave me this UK power strip.

So I just wanted to mention, I talked with Martin about possibly cleaning this thing

up and retrobriting it.

But he decided he wanted to keep it original due to the various stickers he has on it.

So, one thing that stands out to me is how dark the phosphor is on this 1084 compared

to the ones I've seen here in the united states.

This one is almost black by comparison.

Here's another problem I ran into.

This monitor has a scart connector on the back.

Nothing ever used this connector in the USA.

And so there's nowhere for me to plug in my LCA wires.

Fortunately, it does have composite video, so I'll have to use this.

And yes this particular cable uses red for composite.

And it does appear I can also use the RGBI cable for the 80-column display so I can test

that too.

OK, I'm going to power it on for the first time.

And this is what I get.

Yeah, there's definitely something wrong.

I tried rebooting it in 80 columns mode to see if there was any better luck there.

But nope.

Nothing.

So, I began the process of disassembling the unit.

This will be my first time looking inside one of these European 128Ds.

Interesting, so the 1571 disk drive inside has its own separate board, which looks pretty

much exactly like the one you'd see inside the dedicated disk drive.

So, it's interesting that the RF Shield has this typical slant in front of it.

It's probably the same shield that they used on the flat version of the 128.

The RF shield also doubles as a heat sink.

but for the moment I'm going to clean all of the heat sink compound off these chips

so that I can tell what they are.

OK, now that everything is cleaned up, I'm going to start troubleshooting.

The first thing I usually do is check the voltage.

And we do have a good 5V reading.

But when I went to test the 9V AC, I found basically nothing.

So, ideally I'd like to plug in my standard 128 power supply but mine has a connector

like this, but they didn't populate that connector on this board and instead gave us this pin

header here.

So what I'm going to do is use my NuBrick power supply and this female dongle and that

should give me access to the different wires.

I'll use some of these alligator clips to connect it to the header.

I know this looks questionable, but I checked with a meter before plugging it in to make

sure nothing was touching that shouldn't have been.

OK, time to try it again.

And well, nothing.

So, as I feared there's probably more than one thing wrong with this computer.

Next I decided to take apart my flat C128 so that I could start swapping chips around

to see which chips do or don't work.

And I already know that certain chips probably aren't compatible like the video chip because

mine is NTSC and Martin's is PAL.

And to make matters worse, a lot of the chips I wanted to swap are soldered directly to

the board in my 128, which makes this a far more challenging job.

I still couldn't figure out what was wrong with this power supply and I was feeling a

bit frustrated, so I called up a fellow Texan.

She's probably the worlds foremost expert on the C128 next to Bil Herd himself.

And she was happy to come by.

This is June from the youtube channel Nybbles and Bytes.

She jumped right in and got straight to work.

She also brought some equipment I didn't have, such as this nice oscilloscope.

The first thing she wanted to do was verify all of the clock signals, starting with the

main crystal, which was outputting a proper 17.73 MHz signal.

Next she found the 4 mhz clock, and that was working.

But then we got down to the CPU clock and there was nothing.

Ok, so I have to explain something here.

June and I were both under the impression that this computer would actually boot and

run with just the 5 volts DC.

The Commodore 64c works exactly like that.

If you don't have the 9 volts, but you do have 5 volts, it will still boot.

I think it won't have like audio and I think the user port and cassette port don't work,

but otherwise the machine will still Boot and run.

And so, we thought the 128 was like that too.

It wasn't until June traced down the lack of a proper clock signal to the lack of 9

volts that we realized this machine would never work without it.

That being the case, June turned her attention to the power supply to see if she could figure

out why it wasn't working.

After a few minutes she realized what the problem was, kind of by accident.

She noticed when lifting the board by the 9V transformer that it wasn't really attached

to the board on one side.

Well, that's certainly a problem.

She thought she might be able to repair the transformer so she desoldered it from the

board.

Now you can clearly see where it was broken.

So, I asked Martin how this computer was shipped from the UK to the USA and he said somebody

had actually brought it over in their luggage.

And, we all know how the airlines treat suitcases, so I think it's a pretty safe bet that that's

this was damaged.

June attempted several things in order to make it work again but ultimately found that

the windings seemed broken inside the transformer because no current was passing through the

coil.

So, we gave up on fixing that and returned to my idea of using an external power brick.

But, this time June decided to desolder the the power cable from the old power supply

and solder it to my nubrick.

Now we weren't actually expecting this to fix anything, but figured it would help track

down the secondary problem.

However, when we powered the unit up, the thing just started working.

I have no idea why it didn't work for me when I did this earlier.

However, our celebration was short lived because the monitor eventually went bonkers.

And it was doing this on both the 40 and 80 column display.

We were pretty sure this was a monitor problem, but I looked all around and tried a dozen

monitors and I just couldn't find anything around here that was PAL compatible.

I think a lot of people just don't realize how rare PAL equipment is in the USA.

I even tried a couple of modern LCD monitors thinking that they would support PAL and they

didn't.

However, we eventually got something working by using a video capture device that supports

both PAL and NTSC.

At this point we performed a number of tests, including playing Petscii Robots 128, which

uses all 128K of RAM so we thought that'd be a good test of the RAM.

Two weeks later.

OK, so finding a transformer that fit the volts, amps, size, and pinout proved to be

impossible.

But June found this one that's a close match, at least electrically speaking.

So now the challenge is to figure out how to mount this thing in here.

It looks like mounting it sideways and bolting it to the inside wall may be the best bet.

So, one little observation that I want to make is that this goes inside of there as

you've seen.

and then believe it or not, the way this thing mounts in the machine is actually upside-down.

This is the top of the power supply.

Which means, under normal operation, this board is hanging upside-down.

which would explain why, this transformer broke.

So that means under any heavy jostling of the machine, this power transformer is going

to be prone to wanting to break off.

And that's exactly what happened.

OK, time to drill some holes.

I did check with Martin before doing this, and he said it was fine.

OK, so I've placed all of the original wires back on.

And These wires are for the transformer, as are these here.

All right, lets screw everything down.

I'm going to put a zip tie through these holes, as I may need it to help hold this thing in

place.

To make this fit nicely, I'm going to need to bend these leads.

And here we go, let's bolt this thing in.

It turns out this thing is mounted so strong, I won't even be needing this zip tie, so I'll

pull it back out.

And now I just need to wire this thing up.

The middle two leads are going to get connected together since we're doing 240 volts input.

And the reason I'm not bending these wires through the holes is to make it easier to

get the heat shrink over when I'm done.

And after one final check, I sent a photo to June and she noticed I had made a mistake,

so I am going to need to move this wire over to a different lead.

And there we go.

everything looks nice.

OK, everything is put back together.

I've plugged it into the transformer, but I haven't turned it back on yet.

I'm not going to lie.

I'm actually a little nervous about what might happen.

I've found myself in this sort of situation before, everything looked good, turned the

power on, and well, things went south.

So, anyway, so yeah, I'm going to turn it back on.

Technically nothing should happen.

And then, assuming nothing happens, I'm going to check the voltages.

If something blows up, well, we'll deal with that.

All right, here goes.

OK, good.

nothing happened.

Let's check voltages.

Checking the AC voltage coming out of the power supply I'm reading 12.93 volts.

That's a bit higher than I was expecting, but there is also no load on the transformer.

I checked the specs on the voltage regulator and this should be fine.

The next thing I want to do is add a little extra protection since the connectors on the

transformer are so close to the metal case.

So I've cut a piece of cardboard and applied some mounting tape.

And I'll just stick that right there.

That should work fine.

And now to put the power supply back into the computer.

OK, the power supply is in.

And I've been wondering what this toggle switch was used for.

I had noticed that not only was it not connected to anything inside, but the guts of the switch

were even missing.

Well, I asked Martin about this and he said it was to turn the fan on and off because

the fan was loud.

Well, that makes sense because he was working with music, but I guess the switch was removed

at some point.

OK, here's the big test!

I couldn't be happier to finally see this thing working.

Unfortunately, my excitement was short lived.

I had tried loading a game from disk and found that the disk drive was not working.

The drive acts like it is working, but it can't seen to tell that there is a disk in

the drive.

I tried several known good disks and constantly got the error message "drive not ready."

Remember, when we ran the game earlier in the video, I was using an external SD card

reader for that.

So, I called up Martin and I asked him if he knew if the disc drive worked the last

time he powered up the system.

And he said, actually it's been a really long time since the disc drive was used uh in fact

he said it was very rarely ever used even back in the day.

He said the way they did development was they used an Atari ST development system, where

they would do all of their coding and compiling and it was connected to the commodore 128

through some sort of cable, probably the cartridge port, and they would send the test code over

directly to the computer without the use of a disk drive.

So, he actually said the last time the disk drive was probably even used for anything

was probably around 1989.

So who knows how long this thing has been inoperable.

So let's see if we can get it working again.

I figured I'd run the diagnostic program, which would require connecting an external

drive as drive 9.

In the program, I can configure my working drive as drive 9, and the drive to be tested

as drive 8.

I first ran the logic board diagnostic, which wanted me to verify a steady blinking LED.

So for that, I needed to connect the LED back up to the drive.

After that, the drive will flash the result code on the LED, and a single flash means

the system board is OK, and that appears to be the case.

So the next test I performed was a read/write test.

I used an unformatted disk for this.

And well, it came back and said the system is failing.

Unfortunately, it didn't tell me anything I didn't already know.

I decided to crack open my regular 1571.

I needed to borrow the drive mechanism out of it so I could at least determine if the

fault was with the board or the mech.

Even though the logic board says it is good, there are analog components on there too that

could have failed.

OK, so here's a known-good drive mech.

I'm just going to lay it on top and connect it in.

And guess what?

it's booting.

So, I know the problem is with the mech, but I can't visually see anything wrong.

I did test the resistance on all 4 head coils and they all showed around 10 ohms.

So I think those are good.

I removed it from the computer for closer inspection but still couldn't see anything

wrong.

Two days later...

So I managed to find a local person who had a 1571 drive mechanism, but unfortunately

this one is made by Alps and the original is a Mitsumi.

it's possible it may work, but there's another problem.

The wires aren't long enough to fit the 128D.

I had planned to transfer some of the sensors and stuff over to the new drive so the wires

would be the right length, but I can't do that if the mechanisms are totally different.

Two more days later...

So, I just can't find an exact replacement and I'm sure, you know once the video goes

live, there'll probably be a hundred people raising their hands saying "I know exactly

where you could have gotten the right Transformer and I know exactly where you could have gotten

the right drive mechanism."

But the reality is we looked around for a couple weeks and we just have not been able

to find anything, so what I'm going to do is I'm just going to go ahead and use the

drive mechanism from my personal 1571 and then I think Bo Zimmerman may have a replacement

for me when I go back to Austin next, so hopefully I can get my 1571 repaired later.

OK, so this is the original, and here's the one we're going to use.

On the bright side, the drive head wires are already long enough.

In fact, the new one is even longer.

But these are the ones that present a problem.

So, what I'm going to do is unscrew a number of these sensors, like the track zero sensor,

the index sensor, and the write protect sensor, and just transfer those other to the other

drive.

See, here's a whole bunch of stuff I won't need to lengthen the wires on.

I'll just re-attach all of that to the new drive, and there we go.

So here's where we're at now.

At least half of the wires are now long enough without cutting anything.

But if I swap out the stepper motor or some of these other things, I'll have to re-align

the head, which is not something I want to do.

So I think what I'll do is lengthen these wires, and the way I can do that is simply

by cutting off the wires on the other drive and re-attaching them.

I'm going to try for the first time a new soldering method that has been suggested to

me for joining wires.

And well, I don't know if the joint is any stronger, but it is much easier to solder

because the wires are linked together.

And don't forget the heat shrink.

And now this wire is definitely long enough.

All that is left is to do this one.

OK, wires are all done now.

So, let's re-install the drive.

And now it's time to test it.

And there we go!

It's booting!

I think this computer is fixed!

Time to re-assemble.

After getting it re-assembled I really wanted to get some sort of monitor working on it

for when Martin comes to pick it up from me.

I tried another fancy modern LCD television, but it would only display in grayscale.

I thought maybe I could try the RF connection.

Of course, I wasn't even sure which channel it would be on because analog station frequencies

weren't the same between the US and UK.

But it did find a station.

Unfortunately, the picture was even worse than using composite.

Here's another LCD television from the early two thousands, but it just rolls.

And here's another modern LCD and doesn't even try, just says it is an unsupported mode.

So then I thought, what if I connect an RGBI monitor to the 80-column output.

My understanding is the frequency used in the UK should be the same as the US for this

type of display.

However, that didn't work either.

But I finally got something working.

I had almost forgotten I owned this, but it's called a retrotink, which converts vintage

video signals to HDMI.

And it didn't have any trouble displaying a picture for me.

And a few days layer, Martin showed up to see his computer.

He seemed pretty happy that it was working again.

I did appologize for not repairing his monitor, but that was just going to be a nightmare

for me.

So, maybe he can send it to Adrian or somebody to repair.

By the way, if you'd like to see more repair videos, I've started producing regular videos

over on The Time Rift Arcade youtube channel, where I show how we are repairing these arcade

machines.

It certainly has a lot of overlap into repairing vintage computers, but these machines come

with their own set of challenges.

So I encourage you to subscribe, and I'll put a link down in the description.

But that's all for now, so as always, thanks for watching!