MiSTer FPGA : User Serial Connector to C64 IEC - part 2

This is part 2 of my journey, building my own custom board for C64 peripherals for MiSTer.

If you haven't read part 1, you can do so here Part 1

Today my boards from PCBWay arrived, so straight to the workbench to put it together.

The board I had made actually has two functions the IEC bus and a keyboard interface for a C128 keyboard that I use for my MiSTer.

The new board makes a pretty big difference, here are my previous prototypes that I had hanging off my MiSTer.

What I used to have

My original custom IEC bus for my custom C64 core

 

The new IEC Bus for the official core

 

A C128 Keyboard to USB interface

 

What I have now!

IEC on the Left - C128 Keyboard interface on the Right

Giving it a test run before unplugging my MiSTer

The MiSTer stack and my new board looking neat

From the back with the IO port

All plugged in and working

Make your own?

I know my requirements are pretty niche, but if you want one you can get one 

from PCBWay

 

You'll also need these parts:

 

• Logic Level Converter
• 6 Pin Din Connector 
• USB 3 Connector
• Arduino Micro 
• Switch ( to reset Micro )
• D-Sub Connector ( for c128 keyboard )  

 

Programming the Arduino

I found a great piece of Arduino code for C64 keyboard on the Internet.  

I've hacked it to suit my needs and to drive a C128 Keyboards extra keys.

 

You can get my hacked version of the code from Git Hub.

 

A special note: 

  Once you upload the code it is really hard to re-program the Micro with the keyboard code again.

  Simply upload the blink example code first ( that comes with the Arduino IDE ) THEN reload the 

  keyboard code.

 

 

PCB and Schematics 

 

I have released the board schematics and design here.

https://github.com/yellperil/MiSTer64

 

 

MiSTer FPGA : User Serial Connector to C64 IEC - part 1

I have been using my own custom hacked code and hardware on the MiSTer C64 core for a few years now, but finally it's in the official core.  This is awesome for two reasons, 1) I don't have to keep recompiling my cores and 2) it uses less pins and also has reset.

Initially I was confused about how the in/out pins worked and I also got lost in a mess of wires.  Eventually, I bit the bullet and rebuilt it from scratch and it worked.

You'll note that the uIEC is powered from my USB hub.  The DE10-nano doesn't have enough juice to run it.

I even put together a board for my IEC bus and my USB Arduino Interface for my C128 keyboard. They should sit together nicely under the MiSTer. I hope it works :-) . . . ( see Part 2 )

Here is a dump of all of my notes . . .

Caveat

Be very aware that this plug contains both 5v and 3.3v and that you are connecting a 3.3v system to a 5v system.

a) Wiring this up incorrectly may destroy your MiSTer!

b) You must also use a logic level converter

c) Verify everything I have written and linked to, I am not responsible for any mistakes below.

USB Cables

Please note that the pins of the USB 3 cable ends do differ on plug in some places (not all are straight through).

C64 Core Document

You should refer to the C64 core documentation : https://github.com/MiSTer-devel/C64_MiSTer#user_io-pins

USB 3 Type A Receptacle on the I/O board and DE10-nano pins

I have previously wired my own IEC bus directly from the DE10-nano. However, please note that the official MiSTer I/O board has 100 Ohm resistors between the pins and the USB Connector. The Newer I/O boards (5.6 and up) also have 10K Ohm weak pull ups.

However the IEC bus works fine without the pull ups ( I have a 5.5 board ).

Study the schematic for the pin out on the I/O Board : https://github.com/MiSTer-devel/Hardware_MiSTer/blob/master/releases/iobrd_5.5.pdf

or this one with the pull ups : https://github.com/MiSTer-devel/Hardware_MiSTer/blob/master/releases/iobrd_5.6.pdf

Verify the pin numbering of the USB 3 Type A Receptacle on the I/O board : https://au.mouser.com/datasheet/2/837/USB1075-1948891.pdf

Verify the pin translation of the USB 3 Type a Receptacle on the I/O board : https://pinout.net/pinout-scheme/507/USB%203.0%20SuperSpeed

Make the necessary adjustments / translations for your cable!!

The pins for the USB connector ON THE I/O board are :

1) 5 V

2) RS232 Tx

3) RS232 Rx

4) GND

5) ATN

6) DATA

7) RESET

8) CLK

9) 3.3v

Logic Level Conversion

I used this Logic Level converter : [url]https://www.sparkfun.com/products/12009[/url]

Please read the data sheet for usage.

L side is MiSter side

H side is IEC side

LV1 ATN => HV1 ATN

LV2 DATA = HV2 DATA

LV connect to 3.3v

GND connect to GND to GND ( I powered the Logic Level Converter entirely from the MiSTer so it has a common Ground, if you don't then you'll need to separate the grounds to each power source, Read The Manual ).

HV connect to 5v

LV3 RESET to HV3 RESET

LV4 CLK to HV4 CLK

IEC receptacle

I used this part for the IEC receptacle on my PCB design : SDS-60J : https://au.mouser.com/datasheet/2/670/sds_j-1778983.pdf

1) SRQ ( unused )

2) GND

3) ATN

4) CLK

5) DATA

6) RESET

 

Please see Part 2 for Schematics and Design!

https://8bitshardway.blogspot.com/2021/07/mister-fpga-user-serial-connector-to_2.html 

Atmel ATF1504AS(L) CPLD Development Board - Part 9 - DIP 14 and 16

After replicating the function of a 74HC00 in part 8, I became a little obsessed with creating a programmable CPLD that would fit a DIP format.  So after a little bit of playing about in DipTrace I came up with this.

I created two versions, a 14 pin version and a 16 pin version.  Both use the ATF1504AS(L) but in a TQFP package.  I also used 0603 SMD parts for the Capacitors and the Resistors, which although they are tiny, aren't too bad to solder on by hand. Just make sure you use Flux.

Because the PLCC and TQFP have different pinouts I first needed to find the correct MAX product to write for.  I am using the ATF1504ASL-25AU44-T which translates to this part from the conversion guide.

So I'll copy the existing project and set the part to EPM7064STC44-5.

 

 

I will also need to change the pins to match the new layout

 

 

Now I upload it as a 1504ASL ** NOTE the "L"

I had to make special jig to program the CPLD as used 1mm pitch not 1.25mm.

Here is the comparison between the 74HC00 and the CPLD.

 Here it is working just fine in my Z80MBC.

Building the : Cart64out by Sven Petersen ( kits available )

I've been looking at cart breakouts for quite a while now, but I've not had the skills or equipment to warrant one, that is, until now.  I recently bought and started using a Digital logic probe from Kingst and it it is a wonderful piece of kit.

So with my Kinst probe in conjunction with Sven's cart I am now looking at two more projects:

    1) Cart repair : I have two carts that I need to repair, both of which seem to be fine, but just don't work.

    2) Cart design : I'd like to try my hand at building a cart with a CPLD or FPGA.

Anyhow this build was pretty simple but I think It's going to be really handy.  The parts themselves are readily available apart from the cart edge connector, luckily the vendor was pretty good and I got a personalized deal for 30 units.  This is fine as I'll use them for board repairs and I may sell a few.

This is a bit of a niche project, I have quite a few spare boards and parts.

If you're in Australia and would like one please get in contact with me.  If you're outside of Australia, it's probably not worth while due to postage costs. 

The cost is $30 (AUD) for a full unassembled kit which includes:

• 1 x PCB
• 1 x edge connector
• 1 x 40 pin connector
• 1 x 18 pin header ( you can remove the 5v and GND pins if you wish like I have )
• 1 x momentary switch
• 1 x resistor network
• 2 x standoffs
• 2 x screws
• 7 x jumpers
• Standard Tracked parcel post to anywhere in Australia
  

I'm probably losing money on this I'm too scared to do the sums and find out, but it would be cool to see them other peoples hands.

Repair : Commodore 128 blackscreen repair - part 2

This is a continuation of the Commodore 128 blackscreen repair.

Ohh Ohhhhh . . . We're half way there . . . 

Just before I started recapping the C128 I suddenly had a thought. I had only run the C64 diagnostics, was that really enough? 

It turns out that the C128 mode of a C128 does not check it's memory on startup and the C64 dead test cartridge only tests the 64K of RAM in the LOW Bank and ignores the HIGH Bank.

However do not fear there is also a C128 dead test cart available.  I went to the World of Jani and downloaded an image from the collection he has amassed in this awesome blog post : Diagnostic Carts and Manuals.

Making a C128 diagnostic cart

I downloaded a cart from from the World of Jani on to a 8K EPROM 27C64 and tried it with one of the many VersaZif64 carts that I had made quite a while ago.

No matter what I tried I could not get it to work.  I read Jani's blog on making C128 carts and decided that even though the ROM image was only 4K, a 16K cart was probably required for it to work.

So I tried using the 785260 image on a 16K EPROM 27C128 and it worked first go.  For the VersaZif64 cart you need to set the all DIPs to OFF and then turn the selector to 16K

Where's the rest of the RAM?

Running the C128 Diagnostic Cart it brought up the unseen memory issues lurking in the background.

** You can ignore the other errors, I do not have a test harness installed. **

 

I tried putting RAM dips over the existing dips and it didn't change.  Then I replaced both U46 and U47 and it still did not change.  That's really weird.  I thought maybe the Diag Cart is wrong so I downloaded memtest64 from this article.  However it also told me there was a memory issue.  Interestingly when I loaded it, it would sometimes just start the machine code monitor and on one occasion the  Diagnostic Cart also started the Machine Code monitor when it was running.

This is one of the symptoms of the C128 Kernel ROM being bad ( the low one I think ).  I pulled both the High and Low Kernel ROM's, tested them and they were both completely fine.

My next thought was that one of more of the logic chips had failed.  There are a lot of chips so I thought I'd probe around with my Oscilloscope to try to find the culprit, rather than just trying pot luck replacing the chips one by one.

 

However in the end it was just bad RAM and I replaced the entire high ram bank and it fixed the issue.

Recapping

I find it a chore looking at the motherboards to get a parts list for recapping.  Luckily there's always someone whose compiled a list already ( and may even sell you kits ).

I used the parts list from here https://console5.com/wiki/Commodore_128 but I sourced the caps myself as I wanted to do quite a few machines.

The caps in the machine overall looked in pretty good condition, until I popped open the shielding on the video chips.  One of the caps had a nasty burn mark on it.