First time powerup with most of the important bits installed to check for magic smoke and other bits - nothing on the CS, so 'floating' data bus. 'Random' instructions executed until things stop.
Every 1.6 seconds the watchdog chip pulls RESET low (last 3 leds on the 'address bus' LED array are IRQ, NMI and RESET)
Next to write the address map and then a test boot to twiddle a few pins on a few chips. Once that's sorted, then off to write RAM test routines and a watchdog handler
https://www.youtube.com/watch?v=siHQ06EDJh0
Hi (new-ish here)
Re: Hi (new-ish here)
I have recognized during watching your video clip that on the left side of your pcb their are 2 positions for a TO-220 circuit one equipped one empty. I assume voltage regulators?
If so, be careful: applying power on the secondary side of a volt.reg. might cause a reverse current through the chip (to charge the primary side) - this reverse charge can be destructive!
(You may have placed reverse protection diodes I didn't locate, then everything is fine.)
Good luck in your progress!
If so, be careful: applying power on the secondary side of a volt.reg. might cause a reverse current through the chip (to charge the primary side) - this reverse charge can be destructive!
(You may have placed reverse protection diodes I didn't locate, then everything is fine.)
Good luck in your progress!
Re: Hi (new-ish here)
All going well so far - it's running and I'm just waiting for an eeprom so I can make the 'memory map' work (used a 45ns eeprom so I can efficiently organise the addresses)
Captain klutz has prevalied though - apparently I didn't read the MAX690ACPA datasheet properly and missed the bit where it didn't say you can't connect the PFI input to anything greater than VCC (though it did say 'comparator' at which point I stopped reading)
Oh, and this(attached). Not sure what I was thinking, though a little more logic required on RW and OE given I missed the 'X' in the truth table for writing to the RAM even if OE was high (kinda obvious). CE will always be grounded because I don't want to wait.
Captain klutz has prevalied though - apparently I didn't read the MAX690ACPA datasheet properly and missed the bit where it didn't say you can't connect the PFI input to anything greater than VCC (though it did say 'comparator' at which point I stopped reading)
Oh, and this(attached). Not sure what I was thinking, though a little more logic required on RW and OE given I missed the 'X' in the truth table for writing to the RAM even if OE was high (kinda obvious). CE will always be grounded because I don't want to wait.
- Attachments
-
- oops
-
BigDumbDinosaur
- Posts: 9428
- Joined: 28 May 2009
- Location: Midwestern USA (JB Pritzker’s dystopia)
- Contact:
Re: Hi (new-ish here)
piemmm wrote:
All going well so far - it's running and I'm just waiting for an eeprom so I can make the 'memory map' work (used a 45ns eeprom so I can efficiently organise the addresses)
x86? We ain't got no x86. We don't NEED no stinking x86!
Re: Hi (new-ish here)
BigDumbDinosaur wrote:
From where are you getting these 45ms EEPROMs?
https://www.digikey.co.uk/product-detai ... ND/1133334
and (a one time programmable version)
https://www.digikey.co.uk/product-detai ... ND/1118862
-
BigDumbDinosaur
- Posts: 9428
- Joined: 28 May 2009
- Location: Midwestern USA (JB Pritzker’s dystopia)
- Contact:
Re: Hi (new-ish here)
piemmm wrote:
BigDumbDinosaur wrote:
From where are you getting these 45ms EEPROMs?
https://www.digikey.co.uk/product-detai ... ND/1133334
and (a one time programmable version)
https://www.digikey.co.uk/product-detai ... ND/1118862
That Winbond part has mostly been unobtainable for about a year. Please be careful about purchasing chips from eBay sources. It's all-too-easy to end up with counterfeits.
x86? We ain't got no x86. We don't NEED no stinking x86!
Re: Hi (new-ish here)
piemmm wrote:
BigDumbDinosaur wrote:
From where are you getting these 45ms EEPROMs?
https://www.digikey.co.uk/product-detai ... ND/1133334[..]
Re: Hi (new-ish here)
Hi! Been a while - had some more time to spend on this project, so here's an update! Hope it's useful and interesting!
The V1 board was PERFECT SUCCESS! (apart from a total cockup with the 1M ram chip paging - which I'm blaming getting distracted, and the MAX ACPA chip voltages)
V2 is now hopefully being fabbed by eurocircuits with the fixes and a few more additions such as ethernet and USB host functionality (I figured a keyboard would be useful), as well as lots of funky leds that flash and make things look more interesting!
Downsides are that I'm now a 6 layer board, EAGLEs toprouter will segfault if you try to use it, and I hate the fact that the MAX3421 chip doesn't come in anything bigger than a TQFP sized package. I could have simply added a header for one of the many modules available from ebay, but where's the fun in that?
Mouser eventually sent me some genuine 45ns winbond eeproms - only took 6 months (and I also have the knockoffs from ebay - you can tell the quality difference, the ebay ones have pins as thick as kitchen foil!!
)
It's quite cool having a gigantic PCB board you made which has your name printed on it(yep, cc-by-sa will be uploaded for everyone to point out my mistakes once I've got something working
)
The V1 board was PERFECT SUCCESS! (apart from a total cockup with the 1M ram chip paging - which I'm blaming getting distracted, and the MAX ACPA chip voltages)
V2 is now hopefully being fabbed by eurocircuits with the fixes and a few more additions such as ethernet and USB host functionality (I figured a keyboard would be useful), as well as lots of funky leds that flash and make things look more interesting!
Downsides are that I'm now a 6 layer board, EAGLEs toprouter will segfault if you try to use it, and I hate the fact that the MAX3421 chip doesn't come in anything bigger than a TQFP sized package. I could have simply added a header for one of the many modules available from ebay, but where's the fun in that?
Mouser eventually sent me some genuine 45ns winbond eeproms - only took 6 months (and I also have the knockoffs from ebay - you can tell the quality difference, the ebay ones have pins as thick as kitchen foil!!
)It's quite cool having a gigantic PCB board you made which has your name printed on it(yep, cc-by-sa will be uploaded for everyone to point out my mistakes once I've got something working
)
- screenshot of the board from eurocircuits
Re: Hi (new-ish here)
Latest update since I have a moment to work on this again:
It lives! (in a really simple sort of 'make it do something to test the IO' fashion:
https://youtu.be/ILPir9XlJtY
That's a video of it booting up at 90Hz so you get to see 'stuff' happening
V2 board, couple of 'fixes' applied(nothing drastic fortunately) and it seems to work, added networking and usb (still to test) toying with the idea of fusix given it has 1M banked ram to play with (as well as individual dedicated SRAM for each EPROM, just for fun!)
It lives! (in a really simple sort of 'make it do something to test the IO' fashion:
https://youtu.be/ILPir9XlJtY
That's a video of it booting up at 90Hz so you get to see 'stuff' happening
V2 board, couple of 'fixes' applied(nothing drastic fortunately) and it seems to work, added networking and usb (still to test) toying with the idea of fusix given it has 1M banked ram to play with (as well as individual dedicated SRAM for each EPROM, just for fun!)
Re: Hi (new-ish here)
Hi!
Now on v3 of the board (v2 had an arse-up where I trusted an LBR file for the ethernet chip to be 'correct' when it in fact it had transposed pins). Couple of changes - addition of a backplane connector for a custom backplane, extra flashy lights just because, and adding sdcard bits). Hopefully this should be the final board, runs at 5MHz (though can do to 9, but thats out of spec for the e(e)proms). I cheated on the addressing logic, and it works nicely
Now on v3 of the board (v2 had an arse-up where I trusted an LBR file for the ethernet chip to be 'correct' when it in fact it had transposed pins). Couple of changes - addition of a backplane connector for a custom backplane, extra flashy lights just because, and adding sdcard bits). Hopefully this should be the final board, runs at 5MHz (though can do to 9, but thats out of spec for the e(e)proms). I cheated on the addressing logic, and it works nicely
- Attachments
-
Re: Hi (new-ish here)
Not had much time to do anything since, so in the absence of any progress here's the eagle schematic so far (usual disclaimers) - ideas used/adapted from various sites whilst teaching myself
yes, there are probably lots of better ways to do some things!
yes, there are probably lots of better ways to do some things!
- Attachments
-
- 6502EagleBoard.zip
- sch and brd files from eagle pcb
- (365.68 KiB) Downloaded 114 times
Re: Hi (new-ish here)
Had some more time to play again - found that ram really needs to be gated (as mentioned in lots of places here) and fixed a few missing things like the RTC i2c lines (etc, etc) and a stupid mistake with a7/a8 being mixed up on one of the 'address select' proms so that I now have a 'working' system that I had a monitor running on.
Also decided that it's awfully expensive to put everything on one board and get it wrong!, so it's now split up so if I cock something up whilst learning it's not an entire board that starts to look ugly with lots of patch wires! Here's the backplane (waiting on a load of connectors and the other PCB's to arrive for cpu, io, etc, etc)
Also decided that it's awfully expensive to put everything on one board and get it wrong!
, so it's now split up so if I cock something up whilst learning it's not an entire board that starts to look ugly with lots of patch wires! Here's the backplane (waiting on a load of connectors and the other PCB's to arrive for cpu, io, etc, etc)- Attachments
-