Showing posts with label Disk Drive. Show all posts
Showing posts with label Disk Drive. Show all posts

Sunday, 12 October 2025

Changing SD2IEC Device IDs

This question comes up a lot, so I thought I would compile all the various versions into a single post I can point people at.

Background

All Commodore IEC and IEEE-488 devices need a unique ID to talk to the computer. Most disk drives default to device ID #8, but it is common to have secondary drives as #9, #10 or #11.

This is the number in the ubiquitous LOAD "$",8 style commands.

ID #8 is the most useful as that is the one that most things expect. If you press CBM+RUN/STOP on a machine with Turboload, Fastload or Jiffy DOS, it will load and run the default program from drive ID #8.

(Yes that is a real C64, I just have a modified ROM with SX-64 colours to make it easier to take photos.)

If you do SHIFT + RUN/STOP on a Commodore PET or a TED based machine, this will load and run the default program from drive ID #8.

If you use the DLOAD, DSAVE, CATALOGUE etc. commands on those machines, they will again be using ID #8.

Multi-part loaders will often be hard coded to load subsequent parts from drive ID #8, even if you load the first part from ID#9.

Why Change?

If you have a real disk drive like a 1541, or a built in disk drive on an SX-64 or a C128D or an 8296D, those all start off as ID #8, so would clash with an SD2IEC.

Some drives such as the 1541-II have a switch at the back to change the device ID, but most don't. To change the ID of those, it requires cutting a trace between two pads.

You can later return to ID #8 with a solder blob. People sometimes fitted switches to those pads, although drilling holes in drive cases is frowned upon these days.

It might seem tempting to leave the internal drives or real drives set as ID #8 and change the SD2IEC to ID #9, I would advise against that for the reasons given above.

To make your life easy, set the drive you plan to use the most as ID #8. 

In most cases, that would be the fast, reliable and highly capacious SD2IEC, with only occasional use of the real disk drives to copy files off or to transfer disks to images for use with the SD2IEC.

Still Want To Change?

If I haven't persuaded you, there here are a few methods you can use.

Universal Commands

On most systems, you can type in the following commands to change the ID. There are faster options for specific systems I will cover later.

The first command is the one which tells the drive to temporarily change it's ID:

OPEN 1,current address,15,"U0>"+CHR$(new address) : CLOSE 1

In most cases, you could be changing a drive with the default ID of 8 to be ID #9, so the command would be

OPEN 1,8,15,"U0>"+CHR$(9) : CLOSE 1

(I am not sure why it isn't just U0>9, maybe deliberately to avoid accidental triggers, or simply an error that became a feature?)

This will last until the power is removed or the drive is reset. To make it persist, you need a further command: 

OPEN 1,new address,15,"XW" : CLOSE 1

In the above example, that would be

OPEN 1,9,15,"XW" : CLOSE 1

You can now test this out, but remember you need to use ID #9 for everything now.

LOAD "$",9

We've changed out drive letter by mistake

Should you suffer the common symptom of "ID 9 Regret", you can reverse the process using the same commands with the opposite IDs to change from ID 9 back to ID 8.

OPEN 1,9,15,"U0>"+CHR$(8) : CLOSE 1

OPEN 1,8,15,"XW" : CLOSE 1

Easier Methods

VIC20

If you have a VIC 20 and a Penultimate +2 or +3 Cartridge (and if not, why not?), there is a program on the Utilities menu that will do that for you.

This program will prompt you for the current and new device IDs.

It will that run the same commands as above for you.

Commodore 64

There is a version of the above program on the Future Was 8 bit preloaded SD cards, and in the download pack

It lives in the following location:

-cbm/tools_and_utils/sd2iec-id

Once you find that, you can run it as before to change the ID.

And your SD2IEC is now device ID #9

(please don't forget you did that and take it out of the draw 6 months later and wonder why it doesn't respond to LOAD "$",8 - TFW8b had a run of people at one point trying to return "faulty" SD2IECs where they had forgotten they had changed them to ID #9)

What if I have an internal drive #8?

Ah, if you have a C128D or an SX64, then you can't use the above options as the SD2IEC will conflict with the internal drive as they will both be ID #8. The easiest way, if you can, is to plug it into another computer and change the ID on there first.

If you don't have that option, then you can do it using the type in commands above with a slight trick.

You first type in the following, but do not press return yet

OPEN 1,8,15,"U0>"+CHR$(9):CLOSE 1

You then press and hold the drive reset button on the internal drive (under the flap on the SX-64). That should stop the drive responding whilst you have the button held in.

Then with your other hand, press return and the command will be sent to the SD2IEC.

You can then release the drive reset button and you should have the internal drive still responding to ID #8, and the SD2IEC set to ID #9.

As before, if you want to make that permanent, type the second command. This is going to drive ID #9, so no need for the drive reset button trick this time.

OPEN 1,9,15,"XW":CLOSE 1

What about the SD2PET?

That is slightly different. It is designed to be the only device connected to the PET, so it hard coded to ID#8 to avoid any problems. 

If you have an 8296D (lucky you), you can either change the internal drive to be ID #9, which involves cutting the trace in the white circle next to the "1" on the 8250LP board inside. (I think you can also do it by lifting the appropriate leg of the 6532 next to it). Or simply unplug the cable to the internal disk drives.

Adverts

The new SD2IEC and Fastload Combo from The Future Was 8 bit is available now in four colours for those who are picky about such things.

Tindie

I can still ship worldwide from my Tindie store and US orders are now tariffs paid.

Currently it looks like Royal Mail to the US is working. I pay the 10% tariffs as part of the postage to (hopefully) avoid delays and additional costs at customs. The increased US postage costs cover this.

I have recently built some more Mini PET 40/80 Internal boards, since the Mini PET II is still in development but people keep asking, so here they are.



Patreon

You can support me via Patreon, and get access to advance previews of development logs on new projects like the Mini PET II and Mini VIC and other behind the scenes updates. On there you can see the post you would have got this week if Tindie had approved my listings in time. This also includes access to my Patreon only Discord server for even more regular updates (and me moaning about Tindie not approving my listings in time for blog posts)

Sunday, 20 April 2025

Building a 1581 Disk Drive Part 3

What do you mean part 3? I thought it was finished. You promised it was over.

You had the big bumper bank holiday post last week. This is just one more wafer thin post.

Something I skipped over before is the Samsung SFD-321B is a dual density drive. Not double density. Dual density. That means it supports both double density and high density disks, as did a lot of standard PC floppy drives. There were double density only in the early days, but I am not aware of any drives being high density only, except perhaps the thin ones used in laptops and USB floppy drives.

In 3.5" disk terms, double density is the blue ones, the ones with one hole for write protect. They are 1MB unformatted, 720K PC formatted. 80 tracks, 2 side, 9 sectors per track (or 10 in the case of the 1581 and 11 for the Amiga).

High density is the black disks, the ones with the extra hole on the other side. 2MB unformatted, 1.44MB PC formatted. 80 tracks, 2 sides, 18 sectors per track (22 on the later Amigas that supported HD disks).

My first experiences with floppy disks was the BBC micro. On there, it was important to get the right match of disk and drive. The 80 track heads were thinner than the 40 track ones, so if you had a 40 track disk from a 40 track drive and you wrote to it in an 80 track drive, it might not read back right in the 40 track drive as you had only written to half of the track width, the old data was still present in the rest of the track.

It seems the same issues does not affect the 3.5" drives, the number of tracks and crucially the head width are the same, only the number of sectors. Behind the scenes, there are some physical differences in the way the drive writes to the disk, the coating used, the write current etc.

However, you shouldn't need to know anything about that. If you have a drive that supports high density, it will detect the disk type using the hole and write to the disk appropriately.

(as long as you don't start drilling holes in DD disks or covering the holes in HD disks, both of those might work short term but both are misusing the disk and would not be reliable long term)

But wait, the SFD-321B in my 1581 supports high density. Does that mean I can use high density disks in it?

Well, yes. Yes it does.

But, you won't get any more capacity out of it, because the 1581 does not know about it, so it will be formatted 80/2/10 and you will get the same capacity as if it were a normal disk.

There does not seem to be a way the drive can find out the type of the disk, other than reading the formatted disk and the geometry information stored in there.

On a PC, if you have a HD drive and you put an DD disk in there and try to format it, it will try to format 1.44MB, unless you select 720K from the drop down (when there was one) or use /f720 in DOS.

Both the hardware and the operating system need to support for 720K disks, and it was dropped from the format menu from Windows XP onwards, and is not supported by the USB floppy drive I was using.

It may be possible to modify the 1581 firmware to support high density and get double the capacity. I don't see the point, I have not filled up one of the double density ones yet. But that might make an interesting project for someone so inclined. It should also be possible to reduce the timing of the 1581 for the more modern drive and disks and speed it up, should they wish.

I am just using HD disks because I have more good ones of those, they were produced more recently, so are more likely to be in good condition. It means they will not be compatible with a real 1581, but I'm not bothered, I am unlikely to see one of those (and even if I do, I can still use DD disks, nothing has actually changed).

Unless anyone can tell me otherwise, I can just use those and each track will only be half used. They work fine, just as well as DD disks. It's not like anything is being misused, simply underused.

OK, that's the last 1581 post for now.

Promise.

Well, maybe.

You see, I've got this idea for a track display.......

And I want to revist the disk change signal....


Adverts

I understand now is not the best time to be selling on an American site, but I am from the UK. All products are designed, built and shipped by me, here in the UK.

My listings are on Tindie, but if you prefer, DM me or use the contact email on my site to order direct.

All international orders may be subject to local tariffs, taxes or duties, and handling fees etc. Please be aware of this before ordering.

Minstrels

Have you eaten too much chocolate already? You need a different type of Minstrels.

There are Minstrel 2 and Minstrel 3 kits and accessories available.

You can now get a Minstrel 2 kit for $200. Same as the 1980 pricing for the ZX80 kit on which it is based.

Or if you want ZX81 compatibility, you can get a Minstrel 3 kit for $200

There are a selection of other repair and upgrade parts for various machines, many of which can be seen on this PET 2001 board.


Patreon

You can support me via Patreon, and get access to advance previews of development logs on new projects and behind the scenes updates. These are often in more detail than I can fit in here. This was actually three Patreon posts rolled into one, so a lot got cut out. This also includes access to my Patreon only Discord server for even more regular updates.

Sunday, 13 April 2025

Building a 1581 Disk Drive Part 2

My 1581 clone is now complete and working well. It took quite a while to get there. There were three more Patreon posts covering the progress, but rather than post those individually and have people suggest things that I have already tried, I thought it would be best to roll them all up into one post.

I am sure this will be long enough, but if you want even more detail, check out the original Patreon versions of parts 2, 3 and 4. Please consider supporting me there if you think four of these posts a month is worth a pound, thats 25p a post.

The story so far.....

Are you sitting comfortably? cup of tea? some of the posh biscuits? OK, let me begin.

I have built a 1581 clone, put it together with a working Amiga floppy drive and it didn't work.

I know now some of the problems at this point were because I didn't have any properly formatted disks, but even going back I was still getting "74 Drive Not Ready" errors most of the time.

At this point I will skip over lots of things I have tried to get reliable operation.

  • Two different C64s
  • A VIC20
  • A different drive cable
  • Other disk drives
  • Other drive cables
  • Other Rockwell R6502 chips
  • Some MOS 6502 chips
  • Making W65C02S chips work (long story, didn't work)
  • Other 8520s
  • A 6526 (with the jumper removed)
  • Other WD1770 chips
  • A VL1772 (with the jumper removed)
  • Other ROM chips
  • ROM with Jiffy DOS
  • Other 65526 RAM chips
  • 6564 RAM chip
  • Different power supplies
  • Standing on one leg
  • Several different T shirts and one red sweater

I also tried dozens of different floppy disks, all of which had been tested in a PC and/or Amiga

None of the combinations were reliable.

Amiga Drives

This was frustrating because the drives and disks appeared to work fine in Amigas or elsewhere.

When looking through the 1581 service manual, I found a reference to the part number of the drive actually fitted to real 1581s.

This shows 312550-01.

I have seen references that say that is for the A500 and the 1581, but checking the Amiga 500 manual, that shows a different part number, 312554-01.

Both seem to be Chinon FB-354.

So, I found a Chinon FB-354.

This one is working, and is direct from an A500 I have just opened up.

You can see it is actually stamped with the Commodore part number, 312554-01.

So, does it work?

No.

Well, sort of. Like a couple of the other drives, it works sometimes, but the disk change is not recognised properly. It successfully listed the directory, but when I changed disk, it went back to drive not ready.

There are various jumpers on the Chinon drive. Maybe 312550-01 and 312554-01 are the same drive but the 312550-01 version for the 1581 had different jumpers set?

The only picture I have seen online of a real 1581 seems to have a Chinon F-354E. Maybe that is different again?

I had been waiting for one of those to come up on ebay, but the ones on there are a bit too pricey and I am fairly sure it would not be any better.

I think it is down to at least one of the following:

  • The original drives were jumpered differently for the 1581 vs the A500 and we don't know how
  • There is an error in the schematics the clone PCBs are based on
  • There is an error in the clone PCBs
  • There was a mod to the production units not included in the schematics
  • The 1581 never worked properly in the first place, maybe it was not around long enough to fix?
  • Maybe this is why people bodged drive reset buttons onto them?

Time to look online and see if there are any answers.

Samsung SFD-321B

It seemed people had been having success with a PC drive adapter board and a Samsung SFD-321B jumpered for PC.

I had a look and the only drives I could fine quickly and cheaply were ones that had been converted for use with Amiga. That seemed a good bet as I could try that and if it didn't work, I could convert it back to PC.

I tried it.

It didn't work.

Well, as before, it worked some of the time, other times would just get stuck with drive not ready etc.

This is the drive as supplied, with the mods to work on an Amiga highlighted.

I found a few "how to convert an SFD-321B to Amiga" guides, and followed them in reverse to get back to the standard PC configuration.

I went into more detail in the original Patreon post, but I will skip over it now, for reasons that will be clearer further down the page.

Tim and other 1581 replica builders all seem to be using an adapter board, and Tim included the PCB of the one he used, so I build that up. (Thanks again Tim, Thagim)

This adapts a "PC style" drive to work with the 1581.

There is apparently a known issue with using this with a Samsung SFD-321, it needs an additional pullup to one of this pins that are no longer driven by the 1581 board, so I fitted that before I started.

Time to test it out.

Finally a successful format and directory.

I had a few disks that would not format, but cleaning the head and adding some lithium grease to the worm gear seemed to improve that.

I tested that quite a lot, and it seemed to be working well, detecting disk changes etc.

I used CBM Commander to create a real floppy disk from the .D81 image of the 1581 demo disk.

It took over an hour, but at the end, all was good.

The images have to be D81 files, don't try a D64 as it won't work.

Note that Epyx fastload does not support the 1581 drive, so I had to remove that in order to load files. It is understandable, the code it sends to the drive is designed to run on a 1MHz 6502 with a 6522 wired in a particular way, and here it is trying to run on a 2MHz 6502 with a 8520 wired differently. (this was before I added the Jiffy DOS ROM to the test C64)

Not quite plain sailing, but I got there in the end.

That seemed to be working OK, so I started looking to make some sort of case for it, I thought a nice option would be mounting it on a sheet of perspex. All I needed to do was order that, drill it, mount it, make it look nice, take some photos and redo some of the screenshots with the patched ROM clearer display on the C64.

Easy. What would possibly go wrong.....

I ordered the perspex, but it was taking a while to arrive, and so I started working on different things, and no longer had the C64 or 1581 on the bench when the perspex arrived.

I put it to one side to pick up when I had some time and some space.

Building a sort of case

Roll on a month or two and it was now time.

I started by marking out the holes on some tape in not quite the right places.

I then drilled the holes not quite where the marks were, so they will end up exactly where they should be.

Actually, 7 out of 8 did. I call that a result.

(the wire mods on the back of the PCB are from some of the testing I was doing to try to get W65C02S chips to work before I got the Samsung drive, I couldn't, so I have skipped over those and everything is back as it was)

I missed one of the floppy holes by quite a margin, so I stuck one of the feet over it.

That seems fine mounted on some pillars.

OK, that looks good, now I just need to sort out the cabling. I need to run a 34 way ribbon from left to right, and make it look better than this.

I had tried various arrangements, with the drive on either side but it still required a number of folds and did not look great.

I had an idea, but before I started to make any changes, I thought I would check everything was still working.

Oh dear. I was back to getting "74 Drive Not Ready" and various empty responses.

Here we go again, back to the same problems.

I went through the same tests again, and still problems (yes, I tried going back to the standard ROM without Jiffy DOS, and with and without the fastload cartridge, in case that made a difference - it didn't).

OK, let's do this properly

This is getting silly. I decided to have a proper look at what was actually going on with the adapter and the various modifications it makes.

"Drive not ready", that basically means the ready signal is not being set or not being read correctly.

According to the manual for the Samsung SFD-321, this signal will be active (low) when the drive has spun up the disk and it has spun around at least twice.

The problem is, the 1581 doesn't turn on the motor, so it doesn't spin the disk, so it doesn't activate the ready signal, so the drive is 74 Drive Not Ready.

It gets a bit messy because the drive can generate either disk change or ready, and send that to pin 34. The 1581 however requires disk change and ready, so that is what the adapter board must be doing.

It has a 74HCT02 quad OR gate chip on there.

This is arranged as a flip flop to generate the ready signal based on the motor control line and the index hole detection sensor (which pulses once per revolution as an optical sensor picks up a hole in the spinning disk)

It generates a ready signal when the disk is spinning and an index pulse has been detected, and clears when the motor is turned off. Similar to the way described in the drive manual above.

Floppy drive connections

This shows the connections to the floppy drive. I am specifically looking at the Samsung SFD-321B, although most are similar.

First, all of the odd numbered pins on all of these interfaces are ground, so those are not shown on any of these diagrams.

Secondly, the In use, and Drive select 2 and 3 pins are not connected to anything, at either end, so they can be ignored.

There are three sets of jumpers, as explained in the drive manual.

Drive select has two jumpers DS0 and DS1. If you want the drive to be active when the Drive Select 0 signal is active, solder blob jumper DS0, Alternatively if you want to use Drive Select 1, blob jumper DS1. For reasons I will not go into here, the PC and its infamously twisted floppy cables use DS1.

Next is a choice of which signal to feed to pin 34, Disk change or Ready. The PC uses Disk change.

Finally the slightly misunderstood pin 2. This is used as part of the density select for the drive. There are various options for 1MB (double density, 720K in PC terms) and 2MB (high density, 1.44MB in PC terms) and also 1.6MB (?)

The option we need here is OPA and OPB open, 2MB HD if the disk has a second hole in it, 1MB DD if it doesn't. (interesting...... see the next post)

This shows the drive in default PC state.

I had previously concluded that the way to get this to work was to have the drive in PC mode like this, and use the adapter board.

I think it is time to re-investigate that, as some of it doesn't quite line up now.

The general advice online was that a 1K pullup is required to be added to pin 2. Now, this is the pin which is part of the density select option. Fair enough, but hang on, that connects to jumper OPA, which is not fitted, so it is going nowhere. So in this application it appears to be pointless?

It may be required on other drives, or other revisions of the drive, but in this case it is doing nothing. And no, I didn't get it wrong and it should have been pin 2 on the 1581 side, as that is the output of the 74HCT02, which is a strong CMOS drive and definitely does not require a pull up either.

I checked all the other signals, and all the ones that are inputs to the 1581 already had pullups on.

Given that, I decided to investigate further into this and it seemed like I could remove that board from the equation, and try to get the drive working as is.

Looking through schematics of what the 1581 generates and what the drive expects, this is what I have concluded I need to do.

Drive select is easy, DS1 is not connected on the 1581 side, and DS0 is wired to ground, so I need to use the DS0 solder blob.

At this point, I had a wire soldered to the Motor On signal to drive the front panel LED. This is normally lit when the drive is selected, and since DS0 is wired to 0V, it is always selected, so it always on. That is not ideal.

I did look at various options for replacing the LED with a tri-colour one, but the orange was never very good when both were lit, so you can't really tell what it was meant to be showing, so I will stick with the two 3mm LEDs, that is working nicely.

I thought it would be useful to show when the motor was spinning, as it is not a good idea to remove the disk when it is. The Motor On line seemed ideal for that. It is driven low by a 74LS07, so that should be fine to drive an LED.

The anode of the LED is connected to 5V, and the cathode goes via R3 (160Ω) to the controller chip where it gets a presumably buffered version of drive select. I lifted one side and wired that to the Motor On pin.

Now to the issue of Disk Change and Ready. The controller generates both, and the 1581 needs both, so let's give it both.

Ready is selected for pin 34 by moving the 0Ω link to the RDY pads. Disk change is picked up from the pad above, and wired to pin 2, on the pad of OPA, which is not fitted (= automatic density selection).

So that is my completed re-wiring of the drive.

Hang on, that's now the same as the "SFD-321B in Amiga mode" I had before.

Oh well, at least I have worked that through and I am happy that it makes sense.

But does it work?

No.

I did try it without the LED mod, just in case. Didn't make a difference.

In all these cases, when I say it doesn't work, I mean it works sometimes, not other times, and rarely after a disk change.

More things to try

Checking online, several 1581 problems were put down to the power supply. I am using an external 5V supply, which I know is good, and the voltages are reading fine all over the board and the drive.

I do see two electrolytic caps, and I am sure someone is going to suggest changing them, so I did.

They were quite small caps, but there is nothing in the way, so I fitted physically larger ones which is generally a better option. They were 33µF 25V and just across the 5V rail, so the precise value is not important. (this is a 5V only drive)

I went for some nice Nichicon 47µF 100V caps I had to hand, overkill but should do no harm.

Were the originals bad?

No.

Where the new ones better?

Yes.

Did changing them help?

No.

I went back over things again, and reading back over the floppy drive datasheet, I noticed that lots of things were timed from the drive select line going low.

I wondered if it being permanently low might be affecting things?

Then I thought about the Motor On signal, that is low when the drive should be accessed, so could I maybe use that as a drive select?

That also meant I could reinstate the original drive LED, since it would now only light when drive select was low, which was now motor on.

Did....

Did that....

I'm scared to look.

Did that fix it?

Oh, yes, it did.

Great.

This is now the state of the modifications.

Disk change and ready are as before. The LED resistor is back how it originally was, no change there.

Both drive select jumpers are clear, and there is a link from Motor On to the common side of the two Drive Select jumpers.

Completing the case

There is no mod board in place now, the ribbon cable is connected directly to the 1581 board, but still doesn't sit right.

I decided to try a trick from my PC building days, you can reduce the bulk of the ribbon cable by separating it into a number of smaller cables. Using a screw driver to separate it into 4 wire sections (and two 5 way in the centre since 34 is not a multiple of 4), it can then be bundled together.

With a few cables ties in place that looks quite neat, certainly better than the flat cable.

It nicely obscures the alternative resistor placements.

One final touch is a second piece of perspex on the top, picking up on the four mounting points for the drive controller PCB.

That leaves space above the drive and the controller card for some airflow.

Looks rather splendid, doesn't it?

I have now tested this quite a lot again, and it seems to be working nicely.

For best results with an unformatted disk, turn the drive off, and back on, and then insert the disk and proceed to format. It seems to get confused when it does not read correct formatting, and this seems to work with most of the good disks.

I have been using the performance test from the 1541 diagnostic cartridge, that seems to be the fastest way to kick off a format. If it fails, say no to "continue" and then yes to "again" and it goes around again.

When it does succeed, you can let it run through the test. Not sure how useful the results are as the expected disk geometry is different, but it gives the disk and the drive a bit of an exercise and if it passes the disk it is ready to be reused.

I have been using CBM command to move things around between an SD2IEC and the 1581.

Here is an arty shot showing the screen reflected in the perspex on the top whilst copying.

CBM Command has the ability to copy D81 disk images direct to real disks. I was looking for some games that were originally on multiple 1541 disks and had been converted to single D81 images. Thanks to CommodoreBlog for suggesting some.

I had left the Jiffy DOS ROM in the 1581 and with the matching ROM in the C64 it was now running faster. It took about 20 minutes to create a disks, it was well over an hour when I first had this running on the standard ROMs.

That looks promising.

I am not very good at games, but they all seemed to load and play.

Another use for the larger capacity disk is you can put loads of PRG file games on one disk. I put FB64 on first, so you can load that with LOAD "*",8 (or Commodore + Run/Stop with Jiffy DOS).

I have just stared with I few I had on an SD card, including the freeware demo of PETSCII Robots which only has a few levels, so there are only a handful of files that can be extracted from the D64 to fill the disk.

I think the 1581 saga is now complete. It seems to be working nicely now.

Update

It is still working this morning, so hopefully that is resolved now*.

I am not convinced the disk change notification is right, since the motor will not be spinning, but it seems to be working in practice, and I can use the @I command to reinitialise the drive, is press the drive reset button if I decide to fit one.

* Your milage may vary. Things that fix my problem may not fix yours. No guarantee mine will be working tomorrow. Things may be screwed up as well as down.


Adverts

I understand now is not the best time to be selling on an American site, but I am from the UK. All products are designed, built and shipped by me, here in the UK.

My listings are on Tindie, but if you prefer, DM me or use the contact email on my site to order direct.

All international orders may be subject to local tariffs, taxes or duties, and handling fees etc. Please be aware of this before ordering.

There are a selection of other repair and upgrade parts for various machines, many of which can be seen on this PET 2001 board.

There are also Minstrel 2 and Minstrel 3 kits and accessories available.

You can now get a Minstrel 2 kit for $200. Same as the 1980 pricing for the ZX80 kit on which it is based.

Or if you want ZX81 compatibility, you can get a Minstrel 3 kit for $200


Patreon

You can support me via Patreon, and get access to advance previews of development logs on new projects and behind the scenes updates. These are often in more detail than I can fit in here. This was actually three Patreon posts rolled into one, so a lot got cut out. This also includes access to my Patreon only Discord server for even more regular updates.