Commodore 64 Repair and Restore 250407

I picked up an old Commodore 64 last December. When it arrived it looked well maybe not great, but mostly clean.

You can see it still has part of a keyboard cover unit attached. That was quite difficult to remove actually without causing damage..

Internally it looked fairly clean as well. I took the computer out of the case and cleaned the case mostly with soap and water as well as an old toothbrush. I did use some Baking Soda and a bit of water to make a paste like cleaner to scrub some stubborn spots. The case ended up looking pretty good, it is browned a bit but overall not too horrible.

Here I have cleaned the case and keyboard.

Unfortunately the above images are what I was getting out of the computer when I first connected it up. It was flipping through a lot of garbage, I couldn’t even make it out. The camera picked up these images though showing me that something was working. Based on these images I expected a likely issue with the PLA to start. I ordered in a new modern PLAnkton chip to replace it.

Here is the board with the new PLAnkton PLA replacement, and a few caps replaced.

Here she is working after putting in the new PLAnkton.

The computer was working again after replacing the failed original PLA with a new PLAnkton chip. I had to desolder the original chip and solder in a new IC Socket. Then the computer powered up normally except an issue with the Keyboard. It acted like there was a stuck key, after disconnecting the keyboard it still acted like a stuck key. I pulled the left CIA chip there in the upper left corner. It is the CIA that controls the keyboard input. With it out the key wasn’t showing pressed. It was replaced at some point in the life of the computer so it was in a socket. So that seemed to mean the CIA chip was bad. I ordered a replacement chip. As the CIA was super tight in that socket I decided to replace it before putting in the new chip. I actually happened to have a new old stock socket in my stash of the proper size. I installed it and figured, I would test the CIA and see if I could do anything with it. The CIA fit better into this new socket, and it worked perfectly as well. So now I had a fully working Commodore 64.

I did do additional changes, such as replacing all of the electrolytic capacitors. Removing the metal shield around the VIC chip, and installing heatsinks on many of the chips. I also put heatsinks on the Cassette transistor and 12 Volt regulator. I replaced the heatsink compound on the 5Volt regulator. I purchased a full capacitor replacement kit from console5.com.

I also purchased the heatinks, new Din plugs for the Power Supply port, Serial Port, and Monitor ports from Console5.com. I used a new plug for the power port make myself a new safe power supply for the C64 (See my other post on that). I used one of the 6Pin plugs to make a Serial test plug for the Diagnostic Harness (see my other post on the Diagnostic Harness) as well as another for a Pi1541. I picked up the Horseshoe Din for the Monitor/Display Port and made a new Composite Video cable with the audio going to two connectors so I would have copied “mono” audio on my TV. I will be making a S-video cable out of another connector at some point provided I come across a display with S-video on it.

It seems a bit across the board as to “replace the Electrolytic Capacitors” or others will say “Don’t”. I will say if you aren’t using good quality capacitors, don’t waste your time, the old ones may be better or last longer if the new ones are junk. If they are leaking everyone says replace the leaking ones. I say if some are leaking, then others likely could at any time. So far as my experience went on the Commodore 64, I did find that at least some of the old capacitors were giving unexpected values. The old 10uF capacitors varied from 10.8uF to as high as 15uF with 6-8.4% v loss and 3.1 to 25 Ohms ESR. The new 10uF varied from 10.5uF to 10.71uF all at basically 1.2% v loss and 2.9 to 3.2Ohms ESR. I feel at the least that 25 Ohm ESR one was likely to become a problem, I am not so sure about several others. I have to say as the Capacitor kits at Console5.com are $5 to $8 (depending on the board revision) for good brand name capacitors it was certainly worth it for me. They carry them for all of the Commodore 64 Revisions, as well as the Commodore 128, the Commodore 128 Power Supplies (well at least two variants) and the 1541 drives. They also have the Commodore 64 Saver kit.

The keyboard was quite clean overall. I believe the cover was on it most of the time it was out of the box considering it was nearly pristine and only a couple of the springs had any rust on them. I took all of the key caps off with a key cap puller. I then hand washed all of the keys with soap and water. I brushed off the keyboard base and cleaned it with some
91% Isopropyl Alcohol as well, but I didn’t have to take the keyboard apart beyond that.

Black Screen failure..

Shortly after replacing the last of the capacitors it decided to act up and black screen on me. I ran the dead test cartridge and it told me the one ram chip was bad. First I retested my new power supply that it hadn’t went into an over voltage state. It was perfectly fine. I ordered in a replacement ram chip. When I went to desolder the chip, I found the one leg had been clipped off (the VCC pin) at the top and the soldered back on, as well as one of the other ram chips. The only chips that ever seemed to have been changed out was the left CIA and the VIC. I guess it was having an issue and someone was trying to troubleshoot it, I am not sure if that was back when the PLA originally failed or some other point. Anyways I finished removing that ram chip, being careful not to have that leg fall off. I then installed a new socket and put in the replacement Ram chip. I still had the black screen. I pulled the left CIA just in case, as it was socketed and not required to get a display. I then tried pulling the VIC and used contact cleaner on it and on the socket for it. Finally I decided it was time to desolder the SID, I really didn’t want to do that because there was a risk in ruining the chip in the process. I didn’t have anything else to try at that time, short of desoldering more ram chips, which I didn’t have replacements for. I did have sockets for the SID chip as well. When I went to desolder the SID I found the leads weren’t clipped at the factory, and some were folded over. The one lead was touching an adjacent trace. I believe that was causing the fault. I had gotten that far and decided to desolder the SID anyways. I removed it and tested the system, and it came up normally (well normal for no SID or left CIA. I soldered in a new socket and retested, the system still operated fine. After that I went back to the SID and finished straitening the legs out and installed it into the new socket. I was so glad to find it still worked after that as well. I then reinstalled the CIA and swapped back in the original Ram chip, it also worked. After that I ran the diagnostic cartridge with the full test harness through quite a few cycles for good measure. I finished installing the Heatsink on the VIC as I had removed the metal case around it. The VIC is quite hard to get an IC Puller in at with that case nearly touching the bottom edge of it. I actually removed that case before pulling the VIC. Later C64s didn’t have that case over the VIC so I figure I don’t really need it there.

Completed repair, capacitors and heatsinks.

Beyond that I cleaned the mainboard with an ESD Safe black brush to get the loose dirt off and some 91% Isopropyl Alcohol to get residues off of it as well as a bit of flux from the various new sockets as well as around the cartridge and power switch and the various ports.

It was fun getting this old machine going. I didn’t have one back in the day, although I had a used TI 99/4a at some point in the early 90s. So I guess that was my first computer, I didn’t really get to do much with it back then as I had nearly no software for it. I always liked the Apple IIs and such we used in school, and remember writing basic programs for them as well as later on doing that with our 486DX2 66mhz with Dos 6.2 and Windows 3.11 for Workgroups. I worked on the old 80s era IBMs including a few 5050s I had in the late 90s. I guess now I wish I had kept them. I even had a working luggable IBM 5155.. Well I never thought they would be worth anything more than scrap. Still I don’t know that I would have done anything with them. I worked on the IBM PS/2s in school, but never had one, even then I thought they were “weird”. The Commodore 64 is a fun system, and reasonably easy to work on.

It is a nice change working on these old computers compared to working on “modern” computers like I do during the day. It is nice getting into component level troubleshooting and repair.

After getting a second C64, I have found this unit has a only partially working SID. The third voice is not working on it. I only noticed it due to the other C64’s SID playing back more sound on the Diagnostics cartridge tests. If I used it more, I may have noticed some odd audio, but not being familiar with these computers back in the day, I don’t know how long that would have taken. I am figuring on getting a replacement, but I think I want to go with an ArmSID. Maybe next month I will be up to ordering one and see how it goes. My 128 has the same SID model in it, I could check it out to swap it, but I do currently have a fully working C64. I don’t have a good feeling for the chances of picking up a fully working original SID certainly a price that is much better than an ArmSID or other modern replacement. I like to keep them original as much as I can, but these chips are old, and not being made anymore, I don’t expect them to keep running another 3 decades.

Commodore 64 Test Harness

When trying to get my Commodore 64 up and running, I ended up buying a combination Dead Test and Diagnostic Cartridge for it. The Diagnostic side of the cartridge uses a special test harness to test out the operation of the various ports on the C64. I figured I might as well make a full test harness for it. It was a fun project to me, but I am sure there are others out there that would rather buy a completed harness. There are several people selling them. I found a nice modern made one on Ebay for a fairly reasonable price. There are others doing it as well as there is someone on Lemon64 selling them.

I am not including schematics and such, but I have posted the link to the website for the schematics I based my harness off of at the bottom of this post. I will say the Joystick Test portion of the circuit is hard to follow on the schematics with how they are setup. It really looks far more complex than it is. I color coded and traced all of the lines back across the two sheets (one of the big draw backs is the two sheets issue itself), and then I was able to sort out what was going where. The image of the User Port harness does have the layout I sorted from the originals, simply because I had made it and had it available when I took the picture. I did make some changes additions such as, 4 indicator LEDs that indicate power at 4 of the ports. I didn’t put an indicator on the User port although you could. The Serial port has no power pin so it can’t have an indicator. I also moved the joystick test circuitry to the Cassette Port which you will see later on in the post.

The easiest was the Serial Loopback, so I started there. It is just a couple wires on a 6 pin Din plug. I did put a loop of heavy wire on the casing so it could be used to aid in pulling the connector if needed, or to hang it up or such.

The next was a loopback for the Keyboard. The instructions I found for it included that you could make it out of an old IDE Ribbon Cable the earlier 40 wire model. Unfortunately I didn’t have any of those and only had an 80 wire type cable that they said can not be used. They were right the 80 wire type won’t work without some changes to it. The reason it won’t work is there is a center set of contacts that creates links between various pins. I took the connector apart and removed the center strip that was doing that. I then took a bit of the wire (from an 80pin IDE cable) and did the loopbacks directly on the connector.

It worked for the testing like that, but I decided to go back and add a LED and resistor to the loopback, as was mentioned on the Tynemouth Software site. This is good in my opinion, because if you have the keyboard loopback attached, you almost certainly have disconnected the power LED and have no indication there is power in that board. I put the connector back together and attached a pull loop on it to make it easier to remove. I also put a pin into the key pin so it couldn’t be installed incorrectly.

Power LED. There are only 1 wires attached to the LED. The extra wires are only there for extra support but do not go to the LED.

Then I did the User Port Loopback. It is just a series of wires on the pins once the components of the Joystick Test are removed from it. On the original Test Harness from Commodore it was put in what looks to be a modified “cartridge” case and it did have wires going from the Cassette Port and some 4066 ICs to test the Joystick Ports. In my situation I decided to put the 4066s and the Joystick Cables into the Cassette Port test unit instead. It was closer to the joystick ports, and they needed a data signal from there anyways. I also had a nice metal case available to build the board and install it all into. Initially I had installed a wire as a pull assist to remove the User Port Loopback. The connectors I picked up for the User port (and Cassette Port) are very very tight, so that wire did not work very well. I purchased a DSub Case for a 37pin DSub port (sizing chosen by a reference I found saying they fit into them, The Cassette port unit can alternately be mounted in a DB15 enclosure.) and installed the User Port Loopback into that. The DB37 enclosure gives better leverage to let me pull it strait back, it also does prevent it going in as deep, but it still makes good contact.

Original configuration with the pull wire.

Here I have removed the wire and put it in the DSUB enclosure. The line indicates Pin1.

The bottom of the User Port Loopback is Labeled with Red to help point out it is upside down.

The last part of the harness was the Cassette Port and Joystick Ports. The Cassette Port is fairly simple itself, but it actually sends a signal over the 4066s for the Joysticks. Looking at the schematics, I found it rather intimidating trying to figure them out. Being that it was two pages and the “joining” connections were flipped from one page to the other that didn’t help one bit. I am very very glad they are available, but that seems like a an odd choice to make. The 4066s have 4 sets of “switches” in them basically. When a signal is sent to that switch via the Cassette port the switches close and create a connection between the two wires going into it. There are 5 of those switches used by the test harness. The Up, Down, Left, Right and Fire buttons of the two joysticks are wired to each other when those are active. So Joystick 1 fire is wired to Joystick 2 fire and so on. The Paddle pins are each wired through a 120k resistor to 5Volts to give a known voltage on them apparently. I also added a Power LED to the Cassette Module, that shows the power is there on the Cassette port, as I am using the power from it for the 4066s. I decided to add Power LEDs for Each Joystick port as well with Yellow LEDs. They show that each Joystick ports are getting good contact on their 5Volts and Ground pins. The original harness doesn’t test the Ground pin on the joysticks, and if it wasn’t there they wouldn’t work.

Here is an internal view. The “blue” things are 1.5k old stock precision resistors.

Here is the completed unit. Don’t mind the Mislabeling. Note the Power L1 and L2 LEDs. L1 and L2 show that the 5Volts is working on Joystick Port 1 and Joystick Port 2. I thought of putting a another LED at the Input hole again and making it so that it lights when the Joystick ports get their signal from the Cassette port during testing. I doubt I will get around to that though, I am a bit curious what it may look like though.

I also put the Test Cartridge into a 3D Printed case I purchased from ibuy24 on Ebay. It has a 3d printed label plate and the hole in it that lets you toggle the mini switch to change the cartridge mode.

Here is a full test of the harness shortly after completing it. I didn’t have the case for the User Port unit yet.

Overall this required proper card edge connectors, 2 4066 ICs, some decoupling capacitors, a number of resistors and 4 LEDs. It also used a 6 Pin Din for the Serial and a strait through 9 pin Serial Cable that I cut in half to make the joystick cables. Beyond that I used a protoboard for the 4066 joystick circuitry and a salvaged metal enclosure, then the 37Pin DSUB cover as added later.

I based my test harness on the the schematics at Peter Schepers’ webpage http://ist.uwaterloo.ca/~schepers/diagnostic.html

I also took some ideas from the post over at Tynemouth Software. Such as the Keyboard Power LED. http://blog.tynemouthsoftware.co.uk/2014/10/commodore-64-diagnostics-test-harness.html

For supplies I purchased the 6 pin DIN plug at http://www.console5.com where I picked up all my DIN plugs heatsinks and Electrolytic Capacitors for my Commodores. I purchased the card edge ports from China via Ebay, although they are quite tight, I think there may be an alternate one that is made for “thicker” boards? I picked up the 4066s via Ebay as well although they can be sourced for many electronics suppliers. I purchased the DB37 DSUB Cover at http://www.unicornelectronics.com they should have 4066s and such as well I expect. I don’t know if they have card edge connector plugs or not in the proper sizes. I picked up the DB37 cover there when I ordered new ram for my Commodore 64 as well as some other 74 logic chips and various other bits.

Commodore 64 New Powersupply

I purchased an old Commodore 64 Breadbin style computer in December, and it didn’t come with a power supply. That is not exactly a bad thing as the old power supplies are responsible for killing a lot of C64s over the years. When those supplies fail, they often start sending to much voltage into the 5Volt line burning out many of the chips on the board. To see if it made sense to try to bring this old C64 back to life, I hooked it up with a 9Volt Ac Transformer that I had in my spare parts, and my regulated variable power supply. It sort of worked, it did need repaired, but that will be a post for another time.

I did manage to repair the Commodore 64, so it was time to make a proper and safe power supply for it. I took my 9Volt 1Amp Transformer (well 9.5Volt ) and picked up a new Mean Well RS-15-5 5Volt DC 3Amp power supply. I also picked up an enclosure, 4 conductor wire, and finally a proper DIN power connector from console5.com. The other items I had around, the IEC power jack, power cord, fuse holders etc.

I cut an opening in the back of the enclosure for the IEC power port. I also drilled the front for the outgoing power cord and installed a rubber grommet to protect the cable. I used my small drill press to start the holes then for the IEC being obviously not a circle, I trimmed out the rest with my Dremel and a hobby knife. The material the case is made of is reasonably soft and not too bad to work with. Just don’t try to drill or cut to fast, let the tool do the job.

The bottom of the case had some standoffs for mounting. I cut a piece of raw circuit board material to fit into the bottom. I then drilled holes to put screws into it to hold it to the standoffs. This let me have a solid base to attach the heavy transformer and power supply to without putting holes through outside of the case itself. I then mounted the Mean Well supply and the Transformer to the board and installed it into the case.

I really should have the mesh/screen grounded, that is not shown here though.

In the top I mounted a fuse holder, and later a power switch beside it. All of the exposed connections were sealed with Liquid Electrical Tape on the IEC port and the little bit at the base of the Fuse holder that wasn’t covered by the heat shrink for added protection if someone ever opens it up. I also cut out a fairly large opening for ventilation. I figured if I was going to open it up, I might as well make it worthwhile. I had a piece of screen from an old computer case and cut it down to fit and shaped it to bulge into the opening. I don’t know that much heat will be generated, but it certainly should have enough passive cooling going on with that much ventilation. The grill is pretty small, so the odds of anything falling in should be minimal. Also with all of contacts well insulated it shouldn’t be a problem. Something could get in the screw terminals on the Mean Well maybe. To put in the grill, I started by drilling in the corners for the opening with a Forstner Bit with my drill press to make the rounded corners of the hole. I then cut the rest out with the Dremel to get it close and then a utility knife and such to try to keep it clean and strait. It is slightly out of square, but still looks pretty good overall. Of course I did test the supply before attaching it to the Commodore 64.

Here is the completed supply after I added the power switch.

Here is the supply after I finished assembly on the Commodore 64.

It was a relatively simple project. It also was not very expensive overall. The Mean Well was pretty cheap I believe $10.00 with shipping. The next was the case which was around the same price. The 4 conductor wire I used is not very flexible, but even with what I purchased it ended up being more that $20, although I do have a good bit of wire left over. The wire is a stranded alarm system wire, it is only a couple strands and not very flexible. I would also probably crimp badly if it is wrapped up tightly. I figured if the wire becomes an issue, I will end up replacing it with a more appropriate wire. Overall I may have not saved all that much money making it myself compared to the cost of some of the units available theses days, but it certainly means more to me knowing I made it myself. There is not a Commodore 64 Saver circuit in it, but the Mean Well supply does have protection built into it. I did purchase a Commodore 64 Saver kit that I thought of possibly adding to the unit, but I did not have enough space in the case to get it in there. I did later build up the Saver kit, but the trigger voltage drifts with temperature changes. I don’t feel confident the Saver circuit would be all that reliable.

I’ve built a second Commodore 64 Power Supply and done some minor changes/additions. It is a near direct copy of the first model, with a few corrections to make some items easier or better, fuse placement etc.

Commodore 64 New Powersupply 2

1541 two to one

I recently picked up an old 1541 Commodore Disk drive for my Commodore 64. It was a bit ugly so I cleaned it up on the outside and inside before testing it.

I started by taking the case off of course as seen in the pictures. First I used some WD40 to break down the adhesive on the sticker on the top. Then I tried some Isopropyl Alcohol on the permanent marker, but it did not do much. I guess the marker was protected by the dirt.. I went on and used water and regular dish soap with paper towels and a toothbrush for the grills. From there I went to baking soda with a little water to make a paste to scrub the various black marks and the permanent marker with an old toothbrush. It didn’t get all of the marker but it make a big improvement. I then washed it all down with some dish soap again to get any remaining Baking Soda off the pieces. Once I dried it off, I took 91% Isopropyl Alcohol to the remaining marker, and it was able to take care of it just fine. The case is quite browned from the sun, but I don’t know if I will do anything about that.

Internally it was quite dirty too, but I did not think to take any pictures of it. Here is what it looked like cleaned up.

It looked far better and seemed like it may have been in working order. Unfortunately after I put a disk in it was unable to read it, although it attempted to. It turned out that the read write head of the drive was burned out. This is a very common failure on these Newtronic Mitsumi based drives back in the day due to corrosion in the heads or something. Based on the IC dates on the board it is a late 85 or maybe 86.

Above is a picture of the bad head pulled from the drive mechanism. The unit seems to operate normally in every other way, but new heads just aren’t available. With the head being a common failure I considered the odds of getting another Newtronics Mitsumi drive that had a good head as being a big risk. So I looked around for another 1541 listed as needing repaired that had the ALPS drive mechanism instead.

Well we need parts so..

The parts drive came in, it was reported as not working. It made a noise when plugged in but no lights on the front. I looked at the label here, and being in the US the “220V” just won’t work here. The drive wasn’t getting enough voltage to power the 5Volt electronics. It was getting enough to run the 12Volts for the drive motors though atleast to some degree. There may have been nothing wrong with this drive other than that. Still I picked this up for parts. The drive unit itself actually, as this is the ALPS style unit. The drive units are fully compatible, and it is just a matter of pulling the unit and putting it over into my other drive that had the bad head for the Newtronics drive.

220V on the left 117V unit on the right.

While I was in there I did look at the transformer sections. For the 220V there is a “red” tap connected to the fuse, and for the 117V unit the black wire is connected to the fuse. Both transformers are almost the same model number, and both have red and black taps, with the extra tap insulated. Even though they look about the same, minus the last few bits after the dash on the model number, they are different transformers. I did check if switching the Red wire for the Black one on the 220 Volt would change it over to work like the 117 Volt one, but it doesn’t. The only Voltage conversions I have seen on these drives involved swapping out the transformer, basically like I did. The transformer could be changed, also it would be possible to put in and bypass the existing regulators, or supply a proper voltage that can be regulated down by them.

This ALPS 220V unit has the same board model as the Newtronics.

Here is the board in the Newtronics drive, after I had installed the ALPS drive into it.

The boards are the same revision in both drives and appear to have been built just a couple years apart.

Here is the working drive after installing new Electrolytic Capacitors.

The main 6800uF capacitor was dented, so I picked up a set of new Electrolytic Capacitors for the drive from Console5.com. After testing it the capacitors I found one with a ESR of 40 Ohms, so it probably was a good idea to get them replaced. I also redid the solder at the serial ports and a few touch ups of connections that could have been a bit better out of the factory but were ok.

Here is the combined drive assembled. I did basic testing and the drive did read a disk and format a disk fine. I will look to do an alignment check shortly, but it looks like it should be finally working. I have since cleaned up the case from the 220V unit and swapped the converted 117V drive setup back into it. It is less “brown” overall, and doesn’t have a light patch on the top where the sticker was on this one. It did however have a large sticker on the right side, but still looks a fair bit better overall. The change of case doesn’t look significantly different in a picture so I won’t be positing a specific picture of it.

I have also looked over the board from the 220V unit. Someone tampered with it, I don’t think “repaired” is the proper term. I think they realized the 5Volt wasn’t working and removed it and reinstalled it (poorly), they pushed the traces off of the board. I repaired that and a capacitor they had pulled and very poorly reattached as well as one of the jumpers. I also touched up a few of the original solder joints that could use a bit of attention. I then cleaned the board and tested it. The board seems to operate fine, I didn’t do any extensive testing, but is did respond. I figure it is either working fine, or should be reasonable to repair. At this time I am only keeping it for a spare so I packed it up in an anti-static bag for safe keeping.

I am tossing the idea around of turning the extra 1541 chassis into a Pi1541 & Tapuino. I am not sure I feel right putting extra holes into an otherwise good case though. All of the controls would be on the 5.25″ Plate, but it would involve quite a few holes in the back of the case than are there now.

An interesting 1541ii – Cleanup, Figure out and Repair

I picked up this Commodore 1541ii drive this past week When I unpacked it, I was surprised by this little slot with a slide switch in the bottom front here. After opening it I found more oddities.

First these two loose wires coming out the back.

More Wires, an exposed EPROM, and sockets everywhere, and that interesting switch.

Some diodes on a 4 position slide switch.

Here is where the wires from the switch go. To the Device Select Switches.

Here is a close up of the slide switch and that EPROM, where the two wires in the back lead to.

So this thing is a very interesting 1541ii. After seeing the EPROM I figured the wires from it were to switch between some alternate firmware and the factory firmware. It seems originally the wires likely ran out the vent holes in the side and the switch was taped to the side with clear tape. I guess it was removed at some point or fell off. When I turned over the Floppy drive itself, the JiffyDOS label had wedged into it so that sorted that question.

Looks like I found what the EPROM is. It was up in the motor on the bottom of the drive.

The interesting switch in the front appears to be wired as a 4 position Device selection. I haven’t seen such a mod, but it makes sense. This would make for easy device id changes without having to access the little switches in the back of the drive.

The drive also has every chip socketed. That is a lot of trouble to go to. The sockets are various types. I have never seen the odd tall ones before, the blue ones are pretty, but I haven’t had any of those either, although I have seen similar ones. With the standard black ones, several are cut down, which isn’t that uncommon. The chips are quite a mix of dates, the 6522’s are Rockwell chips not the MOS brand ones I expected.

Two other items that were interesting, the pins on the board that the floppy drive connected to were all bent backward a measured bit. Also the metal shield below the drive had triangles cut off the back. These modifications seem to have been done so that the “drive” unit could be pulled out without fully removing the front panel.

All of these things together indicate to me that this drive was very likely from a Commodore computer repair shop. I see no other good reason to socket every chip on the board, and the odd collection of sockets used seems to indicate it was done with what was on hand. I can see where that would be useful in a repair shop to quickly change the device number from the front, otherwise I don’t see much of a reason for that, the 1541ii already has easy to use switches in the back unlike the full sized 1541. The cut shield and bent back pins that made the drive unit easier to remove don’t make much sense unless you happen to be removing the unit a lot, so that may also indicate it as used in a repair shop as well.

I was somewhat surprised about it not being noticeably yellow. It was a bit dirty and had that clear tape and residue on the vent area on the one side. It cleaned up pretty well, although it has a fair number of scratches on it.

Cleaning

It seems to be figured out overall, so I cleaned up the case, drive and mainboard. It was mostly light dirt internally, but there were a few areas on the mainboard that looked to need a bit cleaning.

To cleanup the case I first peeled the old tape off the side. Then I just started off rinsing it down with water. I then used some Baking Soda with a little water on the wet case to make a paste and an old toothbrush. I figured I would work on the scuff marks first and that actually removed the last of the tape adhesive. The 1541ii has a nearly smooth finish unlike the original 1541s with their rough texture. Due to the relatively smooth finish there is the risk of scratching on the case with the Baking Soda, but this case was badly scratched to start, so I was not worried. I would have otherwise used something that wasn’t abrasive to start. While not very abrasive Baking Soda is a mild abrasive. The Baking Soda and water scrub cleaned nearly all of the dark marks from the case. I then washed the case with regular dish soap and a paper towel to get rid of any remaining Baking Soda. Next I dried the case parts.

From there I moved on to cleaning the mainboard. To clean it I used a ESD Safe black brush to knock away as much dirt as I could. Next I used some 91% Isopropyl Alcohol with cotton swabs, and again the ESD Safe Brush to scrub it. That cleaned the remaining old dirt pretty well. There wasn’t much flux on the board, even though the chips had all been socketed. It seems whoever did that cleaned the flux off at the time. There was only a bit around the power switch and serial sockets etc.

I moved on to cleaning the metal shield that the mainboard sets in, it was mostly dust it out and then I did clean it up with a bit of Isopropyl Alcohol. The odd device switch on the shield had some flux where it was soldered on When I cleaned that it seems the one side was only being held by the flux so it started to fall off. I fixed up that and resoldered it back to the shield. I also straitened out the shield a bit where those triangle had been cut off at the back to make it look a little better.

I then cleaned the drive unit itself. It had light dust in it which I removed with the brush. I cleaned the head again with Isopropyl Alcohol and a cotton swab. I also removed the top plate to get all of the dust out from the area the disk slides into.

Repair & Assembly

Well mostly this drive seemed to be in pretty good order except that missing switch and the reattaching of that device switch that started to come off during cleaning.

I had to replace the JiffyDOS switch. For that I used the drill press with a 1/2 inch Forstner bit. These bits do a great job drilling clean holes in plastic cases if you go slow. They can not be reliably used for this kind of thing with a hand drill though. I set the drill press for the lowest speed setting and go very very slowly. I take light shavings and light pressure. You can use a Step Bit, which are actually intended for thin materials like this though. If you go too fast it will melt the plastic and not cut it properly and will risk pushing through and cracking the case. If done properly you get a very clean hole.

Here is the nice clean hole. It may look a little oblong, but that is the picture angle.

Here is my new latching switch for the JiffyDOS mode toggle.

Now that I had the switch installed in the case, I started to reassemble the drive. The drive was missing several screws, so it was time to pull out the spare screw bin and find some replacements. I replaced the three that hold in the mainboard. Once that was in place I soldered the wires onto the new switch. I then used some Kapton tape to secure the wires out of the way so they wouldn’t get pinched somewhere. You can also see that I glued the JiffyDOS label back to the EPROM, I put on a piece of electrical tape first in case the glue didn’t hold well.

Here is a full view of the board. It looks a bit neater than when I opened it up.

Here you can see the board reinstalled and wires managed.

Another oddity is the Paperclip on the top of the latch arm. I couldn’t tell what the intention of it is. It puts very little tension on the arm. I think it may be to prevent the arm from flying too high if the front plastic part is not installed. It may be that it could snap something. I did adjust it a little so it rested right on the unlatch arm without adding any additional force.

Here is a view of the back once the board and case is all together.

I like how the new switch fits in pretty well and is not very obtrusive. I did fire the drive up after getting the board and drive unit cleaned up. It was working to read disks. I didn’t have a proper power supply though, so I just verified basic operation. The JiffyDOS switch also worked. I may post some more on this drive once I get the power supply finished up and do an alignment check and such. So hopefully she will still work then. I don’t have JiffyDOS for my Commodore 64 or Commodore 128 though. I may order it at some point if this Drive works out well.