Modding a TI99 console for DIJIT Avpc usage

[Gary from OPA]

I don't have the original manual that came with my DIJIT Avpc card, or the little jumper wire setup they supplied.

 

But, i briefly remember in my brain memory, i need to lift a pin, connect a jumper and cut a trace and solder one point.

 

The only part I can't remember if in the supplied wire, was there a resistor or something else that connects to one point on the VDP and connects to the pin that lifted out of the socket, I just remember it was wrapped in heatsink tubing, but could it have a small resistor inside it as well.

 

Did anyone recreate that jumper wire, or make a new document that was clearer than the original.

 

I going to try to mod my console tomorrow, and get my AVPC card up and running.

 

I wish I have my TIM/SOB instead, but at least I wrote a better DSR for AVPC so there is less issues with it, and the v9938 at least can support a composite signal until I get my RGB scanner doubler working for VGA LCD.

[JasonACT]

This PDF looks pretty clear, and mostly matches your memory:

 

http://ftp.whtech.com/datasheets and manuals/Hardware/Dijit/avpcdoc.pdf

 

I was wondering though, while you can still read the VDP status register to get the vsync indicator, I thought maybe this device was no longer connected to the TMS9901 to be able to read it "the more reliable" way?  Maybe making it not work with a number of recent demos?

[Gary from OPA]

5 minutes ago, JasonACT said:

This PDF looks pretty clear, and mostly matches your memory:

 

http://ftp.whtech.com/datasheets and manuals/Hardware/Dijit/avpcdoc.pdf

 

I was wondering though, while you can still read the VDP status register to get the vsync indicator, I thought maybe this device was no longer connected to the TMS9901 to be able to read it "the more reliable" way?  Maybe making it not work with a number of recent demos?

Thanks, but their picture does not answer what is inside, the supplied cable, i am wonder how 14 to 15 is connected and to the resistor on the left, is it all bridged together, or is there a diode or resistor inside that heatsink tubing. i going to have to bring up the VDP datasheet and the 99 console schematic and see what is what, their manual does not get into the technical details on what you are doing mod wise.

 

[JasonACT]

6 minutes ago, Gary from OPA said:

their picture does not answer what is inside, the supplied cable, i am wonder how 14 to 15 is connected and to the resistor on the left, is it all bridged together, or is there a diode or resistor inside that heatsink tubing.

Don't, like me, assume anything  but I would have thought pin 14 & 15 being *CSW & *CSR, would want a pull up resistor (10K Ohms?) for each to disable the chip and that cap on the left is just supplying +5v.  So the bulges in the heat-shrink are possibly just resistors, one going in each direction from the middle where the connection is to the cap.

[Gary from OPA]

17 minutes ago, JasonACT said:

Don't, like me, assume anything  but I would have thought pin 14 & 15 being *CSW & *CSR, would want a pull up resistor (10K Ohms?) for each to disable the chip and that cap on the left is just supplying +5v.  So the bulges in the heat-shrink are possibly just resistors, one going in each direction from the middle where the connection is to the cap.

Yes, makes sense. It would need pull-up resistors. I will try that. I think I got some 10k resistors around here. Thanks

[atrax27407]

This should cover it.

 

AVPC doc.pdf

[Gary from OPA]

1 hour ago, atrax27407 said:

This should cover it.

 

AVPC doc.pdf 1.19 MB · 1 download

Yes, same manual linked earlier and I found my original manual as well. But still doesn't tell you what is hiding inside the the supplied jumper wiring harness. Most likely it is two 10k ohm pull-up resistors I going to try that first to be safe see if it works out.

[Gary from OPA]

All that work, and I don't know if it working as I don't have a 6 pin DIN cable. Only 8 pin geneve and 5 pin TI99. Forgot they used 6 pin. Now I need to find a source for it. Ugh

[Gary from OPA]

I found this cable in my junk box in my storage locker it has a different layout of 5 pins looks like it will fit. As you don't need the middle 6 pin as that had 5volts on it. Will build up a cable later on tonight after dinner.

 

[JasonACT]

5 hours ago, Gary from OPA said:

I found this cable in my junk box in my storage locker it has a different layout of 5 pins looks like it will fit. As you don't need the middle 6 pin as that had 5volts on it. Will build up a cable later on tonight after dinner.

 

My thumb can't do that.

 

Edit: You could test if the machine beeps & honks (showing signs of life) without a monitor connected to the card - if you have audio out from the original socket connected to speakers.

Edited January 29 by JasonACT

[+dhe]

Wouldn't it be easier for the creator of the TIM/SOB just to make up another batch?  😃

 

Current EBAY price for 9958's - $45 - $75.

I'm guessing the cost to make a set to be $190-$250 in today's dollars. (Not including labor and shipping).

[Gary from OPA]

@JasonACT yeah my joints in both of my arms and hands can do weird things.

 

@dhe yes. Would be nice to have my own TIM/SOB setup. I will be at some point designing a new version. In the meantime I want to get all my various peb cards working, I basically have at least one of everything ever produced so that helps for software compatibility work, etc. after I get this Dijit up and running, going to get the mechatronic 80col sidecar up and running on my other system.

Edited January 29 by Gary from OPA

[Gary from OPA]

getting there slowly. at least i know my console mod worked, and my dijit card is ok, as i get a ti99 title screen and beep. -- but the DSR is buggy v1.0 - so now to erase an eprom and burn up my 3.02 AVPC DSR and see if it fixes the errors i am having in that, apps lockup randomly, and don't load currently. -- if not, could be maybe my video ram is bad in some way, but ti basic, and editor assembler and loading a file from floppy works, so i think it is the buggy DSR giving me issues with my scsi and horizon and 512k myarc. -- we will know soon enough when i program a new DSR epprom.

[Gary from OPA]

Well i programed the 3.02 OPA version of the DSR, and my system seems to operate better. -- But when i first turn on my TI99 console, the title screen still messed up, until i load like E/A cartridge and do a soft reboot. -- and sometimes i still get locks when loading a program from the Horizon menu. -- but from E/A cart and loading the program directly not using the menu it works fine, funnelweb runs fine. -- i need to get the 80col version and try that out instead and see if there is any more issues. -- still don't know why the TI99 title screen is messed up, the DSR should fix that. -- maybe the console mod not perfect, and there is still issues with the 9918a, i wondering if the two 10k pull-ups is the right answer for pin 14 and 15.

[JasonACT]

6 hours ago, Gary from OPA said:

i wondering if the two 10k pull-ups is the right answer for pin 14 and 15.

There might be too much noise around the motherboard for our guess of 10K resistors, which are quite weak.  Maybe go for mid-strength 4.7K ones, or even strong 1K pull-ups, if the mid ones make it better but it still has an odd error here or there.

 

Also check the trace is properly cut with a DMM.

[Gary from OPA]

53 minutes ago, JasonACT said:

There might be too much noise around the motherboard for our guess of 10K resistors, which are quite weak.  Maybe go for mid-strength 4.7K ones, or even strong 1K pull-ups, if the mid ones make it better but it still has an odd error here or there.

 

Also check the trace is properly cut with a DMM.

Yeah. I going to change them. On mainbyte page about the avpc it says they are 1.1k so I suspect it needs stronger faster pull-up. As lower the resistance the quicker it is driven to high. I check the cut trace that is good, as I get a picture otherwise I  would get nothing as that cut trace insures the vdp 8800/8c00 goes outside the console to the pen.

[JasonACT]

4 minutes ago, Gary from OPA said:

Yeah. I going to change them. On mainbyte page about the avpc it says they are 1.1k so I suspect it needs stronger faster pull-up. As lower the resistance the quicker it is driven to high. I check the cut trace that is good, as I get a picture otherwise I  would get nothing as that cut trace insures the vdp 8800/8c00 goes outside the console to the pen.

Actually, from my understanding, signal speed shouldn't have anything to do with it, these are pulled up permanently - putting the chip to sleep for good.

 

I was more worried some signal noise is allowing them to enable the chip randomly because they are being pulled up so weakly.

[JasonACT]

18 minutes ago, Gary from OPA said:

I check the cut trace that is good, as I get a picture otherwise I  would get nothing as that cut trace insures the vdp 8800/8c00 goes outside the console to the pen.

My Pi Pico PEB side port device has no trouble keeping a copy of what's written to the VDP, it's only VDP reads that don't make it to the side port.  Your video card device would have no trouble displaying what you expect to see if the trace wasn't cut properly.

 

I use this to be able to take a snapshot of the system memory/state using <crtl><alt>print-screen on my unmodified console (having saved the VDP read-only registers, though the full 16KB doesn't get saved this way in my implementation, but it could be).