Jump to content
IGNORED

1090XL remake


kenames99

Recommended Posts

4 hours ago, _The Doctor__ said:

not sure how a voltage doubler would work out at the lower voltages.... I thought he said his cards etc would be 1090 compatible, if it does what the 1090 does and more the it's the 1090e for enhanced maybe....

it would be simple to detect if 12 or -12 were there and act accordingly.

 

Ken

 

  • Like 1
Link to comment
Share on other sites

  • 3 weeks later...

(Mostly) working 320k SRAM card for the 1090...

 

I say "Mostly" because I haven't figured out why SpartaDOS X isn't working with it yet.  RAM testing programs show the memory works, M.U.L.E. works, and DOS 2.5 works with a functional RAMDISK.  But SpartaDOS X doesn't boot.  It's odd.

 

I thought I had it all figured out and this card would be good to go.  When I first installed the card in the 1090 board, I "only" had 64k.  A couple days ago I took a good look at the schematics and realized that one more "undocumented requirement" existed.  Since the 1090 uses transceiver chips, the data bus needs to be active if the memory is called or if the CPU is issuing a write.  Otherwise, the banks cannot be set.  So, I added a formerly unused OR gate into the circuit and now the full 320k is available.

 

I am still trying to figure out the SpartaDOS X issue.  The next version of the card may have a larger heat sink for the voltage regulator, or, maybe I'll just put a small stick-on heat sink on top of the voltage regulator and see if that helps.

 

1259989603_SRAMCard.thumb.JPG.24fd35bf5b78eab39a85832c77c83492.JPG

 

1864866836_600xlwith1090.thumb.JPG.5c3265c8917310c1d9733f48247fbf5d.JPG

 

2144682817_SimpleCheckerMainScreen.thumb.JPG.7ec658cdc81f51225fce4d597e8a9d51.JPG

 

903537736_MemoryBanks.thumb.JPG.90d52fa248d2199a6dc513a4dcfd5284.JPG

 

821738523_ExtendedRAMTest.thumb.JPG.0222f628413d33c8aba05ca5c031630b.JPG

 

 

 

 

1090 w SRAM Card 2.JPG

  • Like 14
Link to comment
Share on other sites

hi,

  I know the programming inside the Side2 cart gets in the way of the programming in the 1090XL, and suspect the same may be true of the hardware part also but have not tested that. The Side carts with the U1M replace the pbi capabilities at least for storage. maybe FlashJazzCat could shine some light on this as far as the Side carts go. until then maybe try an original type SDX cart?

 

Ken

 

Edited by kenames99
wrong info
  • Like 1
Link to comment
Share on other sites

3 hours ago, kenames99 said:

hi,

  I know the programming inside the Side2 cart gets in the way of the programming in the 1090XL, and suspect the same may be true of the hardware part also but have not tested that. The Side carts with the U1M replace the pbi capabilities at least for storage. maybe FlashJazzCat could shine some light on this as far as the Side carts go. until then maybe try an original type SDX cart?

 

Ken

 

I did a little troubleshooting and haven't found any clues yet.  Since I don't have a U1M there isn't any conflict in $D3xx.  A search through the Altirra Hardware Reference Manual didn't result in my finding anything related to a Side3 cartridge that would conflict with a 1090 card.  My 320k PBI card works just fine plugged into the PBI port and is literally what I modified to work in the 1090.  I made a custom config.sys file, on a floppy, with only Sparta, SIO, and AtariDOS drivers loaded.  The 1090 version of the 320k card still fails with SpartaDOS X and the limited config.sys file.  The big difference between the 320k PBI card and the 1090 320k PBI card is that the 1090 320k PBI card has to enable and disable the data bus tranceiver at the right time.  The 1090 also buffers other signals.  But if there was a problem with any of the changes for the 1090 the problem should be a lot more obvious.  

 

I don't have an original SDX cart to try but it's a good idea as that would allow me to see if it's the SIDE3 cartridge or not.  I have been thinking about making a 1090 card that would function as a SIC! cartridge so SpartaDOS X could be loaded on it, though.  I think such a card would have to have a few ROM instructions to disable the internal BASIC cartridge.  Such a card would also have to have the ability to override any other 1090 cards.  (i.e. a memory card)  I'll need to think it through some more.  

  • Like 5
Link to comment
Share on other sites

  • 4 weeks later...

hi everybody,

  I do not know if you would call this history being made but, I have reproduced a working, original Atari designed 80 column video card for the 1090XL expansion box. So after nearly 40 years (35 maybe?) we have true parallel bus 80 column video! I am pleased to finally have it working. Here are some photos of it. It is fully atari design and firmware. having said that, I did notice a peculiarity with it. I had the 40 column display hooked up also, even tho it is supposed to be nonfunctional, after the normal timeout, the 40 column monitor started to do the attract mode (color/luma changes). I will fix that in the ROM code and make that available if anyone would like that done.

 

  I will be putting the schematic and board files on my github as soon as I get it updated properly. thank you all for your patience.

 

Ken

 

20220731_001351746.jpg

20220731_001358674.jpg

20220731_001642600.jpg

  • Like 28
  • Thanks 4
Link to comment
Share on other sites

  • 1 month later...

@kenames99 contacted me tonight and asked if I had any success with the memory card yet.  Well, I didn't at that time.  I found the problem and fixed it.  Now there is a working 320k prototype card for a 1090 (& 1091)

 

To summarize:  The 320k 1090 (& 1091) memory card is almost the same as the external PBI 320k card.  The main difference is it has an on-board voltage regulator and some extra logic for the 1090 (& 1091) data bus transceiver.  The 320k PBI card works perfectly.  But the 1090 version of the same card does not work with SpartaDOS X.

 

About a month ago, I realized that the $D301 decoder was activating for both reads and writes even though only writes would change the bank.  It didn't make sense to me that the PBI card would work, even though it has the same logic, but the 1090 card would not work.  After @kenames99 asked about this I realized that if the data bus transceiver is not on during a read that random information could be written to the card's $D301 registers on a read.  So, I looked at the board schematic, changed the connection for 2 pins, and it works!  

 

Here's a couple pictures.  I used my 1091 prototype because my 1090 board isn't working right.  (Probably a blown voltage regulator due to my lack of adequate heat sinks.  I do need to fix it.)  The 1091 is almost 100% compatible with the 1090 with the only difference being the 1090 has +12v and -12v.  The 1091 does not.  The 1091 also needs a better heat sink for it's voltage regulator too.  But, it is a prototype.

 

Changes to a new card would include bigger heat sinking for the voltage regulator and surface mount capacitors for the flip flop RC circuits.  (I am trying to make the heat sinks an integrated part of the board.)

 

 

 

368989890_320kinstalledcloseup.thumb.JPG.effeaf14e36caae75dece4001807dd3c.JPG

 

946949280_320kwith1091.thumb.JPG.c2ba0a8bbf62c54826029a78b227dc43.JPG

 

 

Edited by reifsnyderb
  • Like 6
Link to comment
Share on other sites

Two questions, the first one for @reifsnyderb: does this mean that 1091-compatible cards will also work with a 1090?

 

And for anyone who may care to answer: would it be possible to get a brief summary of where the project stands at present?  I've been back through the thread but still don't think I have a completely accurate picture of that.

Link to comment
Share on other sites

2 minutes ago, x=usr(1536) said:

Two questions, the first one for @reifsnyderb: does this mean that 1091-compatible cards will also work with a 1090?

 

And for anyone who may care to answer: would it be possible to get a brief summary of where the project stands at present?  I've been back through the thread but still don't think I have a completely accurate picture of that.

Absolutely.  I only came up with a 1091 to have a smaller, cheaper, board.  Anything that works on a 1091 will work on a 1090.  If a card requires +12vdc or -12vdc it will not work on a 1091.  In this case I used the 1091 as a backup because my 1090 isn't working.

Edited by reifsnyderb
  • Like 1
Link to comment
Share on other sites

14 hours ago, reifsnyderb said:

The 1091 also needs a better heat sink for it's voltage regulator too.  But, it is a prototype.

Linear regulators are very inefficient and become even worse so as the input voltage increases. So for 5VDC regulation using one of these you should try to keep the input close to what you are trying to regulate down to (e.g, 7.5-8 VDC for a 5VDC output). I have a circuit I used many years ago that'll give you a regulated 5VDC at up to 2 amps and not require a heat sink what so ever. It's based on a TO style switcher, and I believe there is also a DIP-8 version that would also work with a slightly degraded current capability. It might take me a few days to dig it up, but when I do I'll post what I have.

 

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

7 hours ago, mytek said:

It might take me a few days to dig it up, but when I do I'll post what I have.

 

If you have trouble locating your design(s), I can always post some of my switching circuits.. I always try to use them in my projects, instead of linear regulators.. all SMD devices though.

  • Like 2
Link to comment
Share on other sites

New board revision based upon the current changes.  Hopefully the voltage regulator can be changed out with a switching circuit.

 

Current changes:

 

1.  Larger integrated heat sink for the voltage regulator.

2.  Fixes for data bus logic and $D301 decoding.

3.  All capacitors are now SMD.

 

Edit:  In the interest of encouraging others to develop more cards for the 1090, I've attached the schematic diagram.  Since this card is a memory card, it doesn't follow most of the 1090 conventions.  However, for this application the schematic does show how the data bus should be enabled, how the short phi2 signal is used with R/W, and the decode logic that works with a 1090.

 

421379802_320kcardr5rendering.thumb.JPG.2f8beaf096e82a7c4d5913aed35011bb.JPG

 

schematic.pdf

Edited by reifsnyderb
  • Like 6
Link to comment
Share on other sites

4 hours ago, selgus said:

If you have trouble locating your design(s), I can always post some of my switching circuits.. I always try to use them in my projects, instead of linear regulators.. all SMD devices though.

Go ahead and post your circuit, and when I find mine I'll post it as well.

 

3 hours ago, reifsnyderb said:

New board revision based upon the current changes.  Hopefully the voltage regulator can be changed out with a switching circuit.

You left more than enough space for doing so, because the heat sink will not be required.

  • Like 2
Link to comment
Share on other sites

15 hours ago, mytek said:

Linear regulators are very inefficient and become even worse so as the input voltage increases. So for 5VDC regulation using one of these you should try to keep the input close to what you are trying to regulate down to (e.g, 7.5-8 VDC for a 5VDC output). I have a circuit I used many years ago that'll give you a regulated 5VDC at up to 2 amps and not require a heat sink what so ever. It's based on a TO style switcher, and I believe there is also a DIP-8 version that would also work with a slightly degraded current capability. It might take me a few days to dig it up, but when I do I'll post what I have.

 

hi mytek,

  you are very correct. I even went so far as to suggest people that build these cards should use the switching regulator like those in the 1088xld. that one in my 1088xld never even gets war no matter how long I use it. well, I suggested that over on the 1066 memory card thread. I have the 80 column board working now but the regulator in that gets very hot, I do not trust that. Atari design with that resistor shunt across it is a terrible design. I was running my 1066 memory card and got a voltage spike that killed the card. I have seen thousands of that type regulator circuit fail over my career in service. I am going to release the design as  is because it is the original circuit but I am also going to offer a different design with with a good regulator circuit. any help or suggestions are very much appreciated, from you or anyone else. thank you for your input.

 

Ken

 

  • Like 5
Link to comment
Share on other sites

4 hours ago, selgus said:

Fro my ATARI Portable, I actually use a step-up converter, since I am going from a 3.7V lipo to +5V..

 

This is the device I've used in some of my R2-D2 lighting boards, where driving high amounts of current for the different RGB LEDs..

 

tps54521.pdf?ts=1662133203480&ref_url=ht

hi Selgus,

  that is 1 of the regulators I am considering.

 

Ken

 

Link to comment
Share on other sites

hi everyone,

  I have put the files for the 80 column card on my github at https://github.com/kenames99/80cvc

I did not put the .jed file for the gal there yet, I cannot find the correct one right now. I will get it there as soon as I can. the converted gal code did not work for me. but anyone that wants to get started on a build can do that now, all the gerber files are current.

 

reifsnyderb  is currently working on turning the jed file back to equations.

 

I hope everyone enjoys this.

 

Ken

  • Like 9
  • Thanks 1
Link to comment
Share on other sites

I have new 320k boards being made.  I made the heat sink area a little larger.

 

Also, I checked into the regulator used on the 1088XEL and decided to just make the heat sink area even larger.  The price of the 1088XEL regulator is over $7...if it's available.  Having to add a voltage regulator as well as some extra logic makes this board expensive enough to source the parts for and assemble.  If everything works well on this board I've got a plan to drop the voltage regulator down and make the heat sink even larger.

 

1602175799_320kcardr6rendering.thumb.JPG.c4a362f7f27f72de800b3d0ed5981652.JPG

Edited by reifsnyderb
  • Like 3
Link to comment
Share on other sites

On 9/4/2022 at 4:02 PM, kenames99 said:

hi everyone,

  I have put the files for the 80 column card on my github at https://github.com/kenames99/80cvc

I did not put the .jed file for the gal there yet, I cannot find the correct one right now. I will get it there as soon as I can. the converted gal code did not work for me. but anyone that wants to get started on a build can do that now, all the gerber files are current.

 

reifsnyderb  is currently working on turning the jed file back to equations.

 

I hope everyone enjoys this.

 

Ken

I've got the gal file from a .zip file I believe I retrieved from the Atari Museum files.  It looks like it could be right.  I also think I've figured out how to convert a .jed file back into a fuse map, of sorts, in the data sheet.  @kenames99 also gave me some notes that appear to match the .jed file.

 

 

Link to comment
Share on other sites

I do have the PAL and GAL files provided by atarimuseum.com and I also made a custom .jed for the 22v10. I have a couple that say they are for the 22v10 but they differ from each other and I don't know which is correct. I misplaced the working 22v10 I had in it. I did get hold of 2 real PAL20L10 chips and got them programmed. between me and reifsnyderb we will get this figured out and have that available. thanks for your patience.

 

Ken

 

Link to comment
Share on other sites

While I don't presently have an 80 column board available to confirm this is it, this is a decompilation of the .jed file from atarimuseum.com using jedutil, which comes with MAME:

 

 

Inputs:

1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 16, 17, 18, 19, 20, 21, 22

Outputs:

14 (Combinatorial, No output feedback, Active low)
17 (Combinatorial, Output feedback output, Active low)
18 (Combinatorial, Output feedback output, Active low)
19 (Combinatorial, Output feedback output, Active low)
20 (Combinatorial, Output feedback output, Active low)
22 (Combinatorial, Output feedback output, Active low)
23 (Combinatorial, No output feedback, Active low)

Equations:

/o14 =
o14.oe = i21

/o17 = /o19 +
       i16 & /o20 +
       /i13
o17.oe = vcc

/o18 =
o18.oe = /o17

/o19 = i4 & i5 & /i6 & i7 & i8 & /i9 & i21
o19.oe = vcc

/o20 = i1 & /i2 & /i3 & /i4 & i5 & /i6 & i7 & i8 & /i9 & /i10
o20.oe = vcc

/o22 = i1 & /i2 & /i3 & /i4 & i5 & /i6 & i7 & i8 & /i9 & /i10 & /i11 & /i15 & i16
o22.oe = vcc

/o23 =
o23.oe = i1 & /i2 & /i3 & /i4 & i5 & /i6 & i7 & i8 & /i9 & /i10 & /i11 & i15 & i16

 

 

 

I have confirmed that the "+" sign, such as shown on /o17, is a logical OR.

& is a logical AND

 

After reviewing the board schematic, I think the equations are correct.

 

I've attached the file from the atarimuseum.com website, as obtained via the "wayback machine"

Also attached is a CRT driver disk and documentation.

 

1090 80 col.zip

CRT_Driver_1090_80col_card.zip

 

 

Edited by reifsnyderb
  • Like 4
Link to comment
Share on other sites

  • 2 weeks later...

I received the 320k SRAM upgrade card for the 1090/1091 yesterday.  So I took the chips off of the prototype and moved them to the new board.  It's a success!

 

Everything is now surface mount so those big electrolytic capacitors are gone.  I kept the voltage regulator because other options cost a lot more.  The increased size of the integrated heat sink is working nicely and doesn't get too warm to the touch.  (Integrated heat sinks are practically free, of course.)

 

1604216411_1090320kMemoryModule.thumb.JPG.371c774a99d8a2a26e8bb909b1d99f4e.JPG

  • Like 16
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...