TI Mini Expansion System

[Artoj]

Hi All,

I have found a Full TI PEB Disk Controller among all my electronic rubble but as I do not have a full size PEB box, I opted to redesign the Direct connected board that Lou Amadio made back in 1989. I will be designing  a separate power supply shortly. I have added 2 of the 22x1.27 IDC connector to the back so it can be used to add a small board (still to be designed),  so you can still add any side TI peripherals. Regards Arto    

 

Edited April 10 by Artoj

[Artoj]

Hi All,

I have one lot of my board designs, this will keep me busy for a while, among them are the Mini-PEB Sams 256k/1024k Board, Mini-PEB 64/64 I/O Card, TI99 Port Analyser, 32K mini board, TI99 Direct PEB card, TI99 Joy Port, a Power Supply, 5 x Ternary I/O boards, Light Code I/O, 3 x Parallel port control cards. Waiting for more parts from all over the world and a supply of IC's from Unicorn Electronics. In the mean time I have started a Page on my blog just for TI99 designs and software, I will send a link when I have populated the page. Regards Arto.

 

 

Edited May 1 by Artoj

[Artoj]

Hi All,

While clearing my many boxes of paperwork, I found a version of Peter Schuberts Multifunction PEB Card schematics (32k, RS232x2, PIO, Floppy). While he only released a few working prototypes (1987), it was a functional card, with a caveat that there might be some on going design adjustments. I have started to build a facsimile even with my scant information and vague pictures (bw photo copy). I have added a 74LS251 so the PIO is bidirectional, also two RS232 RA sockets. I remember the DSR had many extras as well, I added a few myself, I have found some room for a possible addition of another function in the version 2 card. Here is the board (25% done), regards Arto.    

Edited May 11 by Artoj

[+Ksarul]

20 hours ago, Artoj said:

Hi All,

While clearing my many boxes of paperwork, I found a version of Peter Schuberts Multifunction PEB Card schematics (32k, RS232x2, PIO, Floppy). While he only released a few working prototypes (1987), it was a functional card, with a caveat that there might be some on going design adjustments. I have started to build a facsimile even with my scant information and vague pictures (bw photo copy). I have added a 74LS251 so the PIO is bidirectional, also two RS232 RA sockets. I remember the DSR had many extras as well, I added a few myself, I have found some room for a possible addition of another function in the version 2 card. Here is the board (25% done), regards Arto.    

Which floppy controller was it using? This could be interesting. . .

 

Is it possible to scan the old schematics? I try to take the ones that are HTF and convert them to Visio to make them somewhat more accessible to the community. I've done quite a few (all posted in one of the documentation threads here on AtariAge), with more to come as I get time to work on them. One of the ones in the repository already is for the Quest 200 board. . .I actually have permission to make more of those (and have a few prototypes that I still need to build/test).

[Artoj]

7 hours ago, Ksarul said:

Which floppy controller was it using? This could be interesting. . .

 

Is it possible to scan the old schematics? I try to take the ones that are HTF and convert them to Visio to make them somewhat more accessible to the community. I've done quite a few (all posted in one of the documentation threads here on AtariAge), with more to come as I get time to work on them. One of the ones in the repository already is for the Quest 200 board. . .I actually have permission to make more of those (and have a few prototypes that I still need to build/test).

Thanks Ksarul, sure at some point all my files will be available, I have some poor photo copies of his hand drawn schematics. If you can identify a chip marked "2889 2446" that would help to finalise my version of Peter's board. The floppy chip Peter uses in all his designs was the 2793, I remember discussing his many choices back in 1985. We worked together on many of his projects, I used to build his stand-alone modems on the weekends, I talk to Peter every week about my redesign of his cards. Before I moved on from the TI, I had the Software and hardware almost built for a TI99 Midi/Audio Music composer unit, I have found much of my old work recently, which I will be posting on my blog. Current design at 60%, Regards Arto   

[+Ksarul]

A 2889 is an 8-pin isolator chip and a 2446 is a 50mV shunt. How many pins is the chip on your board showing and what is it connected to? Beware of some early lots of 2793 Floppy controller chips too. A whole bunch of 1984/1985 date code ranges came bad out of the factory. Later chips weren't problematic (or earlier ones, for that matter), but a lot of companies got burned by the bad ones BITD. CorComp actually redesigned their floppy controller away from that chip after they received a lot of 500 of them for their initial production run and found out most of the chips in the lot were bad. They put a 1773 in its place using a small daughter board socket to avoid scrapping the whole board run.

 

Finding many poorly photocopied, hand-drawn schematics over the years is exactly why I started the documentation preservation project for such things.  

[Artoj]

39 minutes ago, Ksarul said:

A 2889 is an 8-pin isolator chip and a 2446 is a 50mV shunt. How many pins is the chip on your board showing and what is it connected to?....

It seems he had completed his design but gave no context, it could be an 16/18 pin chip, Addresses A9 to A15 (pin 1 to 7) from the 32k Eprom chip and PEB bus, Data D0 to D3 (pin 11 to 14) also attached to the 32KEprom and bus, it also has a CE(pin8) and WE(pin10). I was thinking it might be a set of Cmos Analog gates controlled by CE/WE?? I do think it was necessary, to swap the banks for the DSR the CE is gated by the logic using A3 to A8 and WE is gated by WE on the bus. I looked up this number (2889-2446) and it says a intel chip, but cannot find a data sheet. Peter does not remember much about this design, as he stopped the project at some point, it could have been due to the 2793 chip problem? I am not sure but I remember his system worked fine for years. Regards Arto

[+Ksarul]

1 hour ago, Artoj said:

It seems he had completed his design but gave no context, it could be an 16/18 pin chip, Addresses A9 to A15 (pin 1 to 7) from the 32k Eprom chip and PEB bus, Data D0 to D3 (pin 11 to 14) also attached to the 32KEprom and bus, it also has a CE(pin8) and WE(pin10). I was thinking it might be a set of Cmos Analog gates controlled by CE/WE?? I do think it was necessary, to swap the banks for the DSR the CE is gated by the logic using A3 to A8 and WE is gated by WE on the bus. I looked up this number (2889-2446) and it says a intel chip, but cannot find a data sheet. Peter does not remember much about this design, as he stopped the project at some point, it could have been due to the 2793 chip problem? I am not sure but I remember his system worked fine for years. Regards Arto

I have definitely verified that it is an 18-pin Intel chip, but no luck yet on getting a datasheet. Next stop on my hunt would be a 1980s Intel databook.

[Artoj]

Hi All,

 

I tried to find the 2889-2446 chip (U26) with no luck, so I decided to look at the logic and realised it should be a Ram chip of some sort. I found a CY7C148-35PC that matches the pin outs and is a 1024x4 bit static Ram chip and with the Decoded MBE_ that is on the card using a 74LS138(U3) (DSR >4000-..) it is a I/O location when it gets access via A9-A13 (max 007F) at = >4000 + >001F  = >401F to the CS_. A Nand gate also decodes >5FC0->5FFF so the Eprom gets a CS_ on pin 20. So there is code written in the DSR Eprom that is used to access routines specially written for this card. I have to hunt for for a copy of it now.  From the Data Sheet:

 

Functional Description 

The CY7C148 and CY7C149 are high-performance CMOS static RAMs organised as 1024 by 4 bits. Easy memory expansion is provided by an active LOW chip select (CS) input and three-state outputs. 

 

The CY7C148 remains in a low-power mode as long as the device remains unselected; i.e., (CS) is HIGH, thus reducing the average power requirements of the device. The chip select (CS) of the CY7C149 does not affect the power dissipation of the device.

 

Writing to the device is accomplished when the chip select (CS) and write enable (WE) inputs are both LOW. Data on the I/O pins (I/O0 through I/O3) is written into the memory locations specified on the address pins (A0 through A9). 

 

Reading the device is accomplished by taking chip select (CS) LOW while write enable (WE) remains HIGH. Under these conditions, the contents of the location specified on the address pins will appear on the four data I/O pins. 

 

The I/O pins remain in a high-impedance state when chip select (CS) is HIGH or write enable (WE) is LOW.
 

Here is the card at 95%, where it will stay until I find the DSR Eprom copy. Regards Arto.  

[Artoj]

Hi All,

 

As I have been putting these boards together, I realised that the 64/64 IO board will trigger the sound chip, so I have moved the addressing to >9300 just as Ross Mudie did back in 1988, this will only be a problem if you have the speech synth attached. Here is the updated Schematic. Regards Arto

 

Schematic_TI99-Wire Interface V3_2023-05-16.pdf

Edited May 15 by Artoj

[Artoj]

Hi All,

Over at Classic Computing section there is a thread about the replacement of the TMS VDP chip with a Tang 9k.  I would like to thank retrocanada76 for all his good work, he has been working on a VDP replacement for the Nabu and this one hit the jackpot!! Using a Tang was a stroke of genius, here is my TI99 version 3 without surface mount. Now you can use HDMI or VGA and get V9938 graphics, Regards Arto.

 

 

[Artoj]

Hi All,

It looks like I have to reread the Tang PDF, the input/output might have to be 3.3v? Update coming.

[Artoj]

Hi All,

I hope this fixes the 3.3v <> 5v issue, the LVC is the translator.  Quoted "You must connect VCC to your logic level you want to convert to (say 3.3V), Ground connects to Ground. Wire OE (output enable) to ground to enable the device and DIR (direction) to VCC. Then digital logic on the A pins up to 5V will appear on the B pins shifted down to the VCC logic.". The HC AND gate can run on 3.3V Logic levels. 

 

Now I have to learn about the F18a FGPA files and the Tang 9K to test it, LOL 

 

TI99-Tango V4

 

1    Tang Nano 9K    U1
2    74LVC245         U3,U4
1    74HC08            U6
3    4.7K                 R1,R5,R9
3    2.2K                 R2,R6,R10
3    1K                    R3,R7,R11
3    480 ohm           R4,R8,R12
2    47 ohm             R13,R14
1    4 pin Dip sw      SW1
1    VGA 15 pin        DSUB1

1    Schottky Diode  DS1

3    1N4007             D9,D10,D11

Edited May 24 by Artoj

[Tornadoboy]

On 5/24/2023 at 11:05 AM, Artoj said:

Hi All,

I hope this fixes the 3.3v <> 5v issue, the LVC is the translator.  Quoted "You must connect VCC to your logic level you want to convert to (say 3.3V), Ground connects to Ground. Wire OE (output enable) to ground to enable the device and DIR (direction) to VCC. Then digital logic on the A pins up to 5V will appear on the B pins shifted down to the VCC logic.". The HC AND gate can run on 3.3V Logic levels. 

 

Now I have to learn about the F18a FGPA files and the Tang 9K to test it, LOL 

 

TI99-Tango V4

 

1    Tang Nano 9K    U1
2    74LVC245         U3,U4
1    74HC08            U6
3    4.7K                 R1,R5,R9
3    2.2K                 R2,R6,R10
3    1K                    R3,R7,R11
3    480 ohm           R4,R8,R12
2    47 ohm             R13,R14
1    4 pin Dip sw      SW1
1    VGA 15 pin        DSUB1

1    Schottky Diode  DS1

3    1N4007             D9,D10,D11

I definitely want to build one of these!