CDTV - Setpatching to 3.1?

[Milesunder]

So I have a CDTV that is an earlier production model. S/N is 000236 and is revision 2.2

Not really a big fan of the CDTV but I wanted to do some upgrades to it to make it a bit more usable. He is where the problems have started. I have installed a TF536, RGBtoHDMI and the latest extended ROMs. However, when I try to replace the KS1.3 to 3.1 and above the CDTV will not boot. I have tested all the new parts and have tried various KS roms in the stock CDTV. I have managed to get the CDTV working on V2.05 KS. Interestingly I have some KickRoms from Alinea and when I programmed 1.3 onto the ROM that didn;t work either. One suggestion someone made in the past s that possibly the KS Socket might be damaged and that 1.3 is a 16bit ROM and later ones are 32bit (Not sure if I have this correct).

Anyway with KS2.05 I can boot the machine and it will boot the HD installed on the TF536. So I kinda have a working system.

What I want to do is load at least 3.1 on the CDTV. I don't want to take the machine to bits to see if the rom socket is faulty. I have been looking for a Mapprom soultion but i am not sure if there are any out there that are not super money. So I was wondering if there was a way to boot the system in 2.05 and then it does something similar to what 3.9 does and reloads the 3.1 roms into memory. One thing to mention is that I only have 1mb of Chip Ram.

I am also away there is a softkick ADF that will load a higher version. I do have an external gotek available but I am not sure if this would work.

So does anyone have any ideas how i can do this?

look forward to hearing from you all.

 

[8 Bit Dreams]

Hmm... KS1.3 and KS2.05 are 256KB, where KS3.1 and above are 512KB.
This means that bottom half of the rom is working, and Your problem is in upper half, so probably one (or more) of the uppuer addresslines are shorted to ground.. Grab a Multimeter and check the upper part...I would start with A17:

 

[Milesunder]

Hey 8bit,

I think you mentioned something about this before.

Happy to test. Just need to clarify a couple of points.

The upper part would that be pins 1 to 10 and 31 to 40?

When I find a ground point if it makes a circuit then that is bad. if I get 0 then that is fine and move on?

 

[Milesunder]

So I just tested. I used a point on the disk drive securing point. When I tested it appears that pins 10, 11 and 31 are shorted to ground. They bleeped on the multi meter and then when I tested between them they pinged.

What are the next steps you recommend?

 

[Milesunder]

I have taken the board out to have a look. The underneath looks okay

 

[8 Bit Dreams]

I don't like these Pins at all - please resolder them properly:

also please take a look inside the Kickstart Socket - sometimes traces are damaged there If someone removes Kickstart ROM with a sharp screwdriver..

 

[Milesunder]

Thanks, so of the three pins that were grounding. Should any of them actually be grounds?

 

[levo-ace]

At least two should be at ground .

VSS is ground for the chip .
/E is the enable line ( grounded means that the output is always enabled ) .
/BYTE sets the chip to 8 or 16bit . This is a worrying one , looks like it is set by link2 , according to the schematic .

 

[Milesunder]

So I have removed and resoldered the socket and I am still getting the same issue. So are we saying that pin 31 is grounding out and this is why I am having the issue?

Interestingly it appears I have a jp2 on my motherboard which later revisions don’t have. If I cut the trace would that solve the issue?

 

[levo-ace]

looking at the schematic , I don't think that it will . Does that via by the socket go to pin 31 ?

 

[Milesunder]

Yes it does. I have cut the trace between the two pads that run parallel to the socket and now there is no continuity on pin 31.

Just putting the board back together to see if it works (fingers crossed)

 

[Milesunder]

So it is back together. The good news is I appear to have soldered the new socket incorrectly. And it is still running v2 the bad news is the 3.1 rom I tried didn’t work. I will try a couple more.

Any ideas?

 

[ronaldvd_nl]

Hmm... KS1.3 and KS2.05 are 256KB, where KS3.1 and above are 512KB.

I'm pretty sure that the ROM size is 256KB for the Kickstart 1.x versions, and 512KB for Kickstart 2.04 up to 3.1. So the ROM size would not explain the issue.

Is the 3.1 ROM a burned EPROM? If so, which type?

And which exact software version is the 3.1 ROM? The CDTV would need version 40.63, which is the version for the A2000/A500.

 

[levo-ace]

If the size of the rom is the issue , then you need to be looking at A17 on pin 1 .

 

[Milesunder]

Thanks Levo, In respect to pin 1 do I need trace where it goes to? Any ideas where it goes to?

 

[levo-ace]

It should go to pin 46 on the 68000 ( A18 ) .

 

[Milesunder]

Thanks. So if I put a probe on a18 on 68000 and on a17 on kickstart there should be continuity?

 

[Milesunder]

Just tested pin 46 on the 68000 and pin 1 on the KS socket and I got continuity. Is this to be expected?

 

[levo-ace]

Yes that is correct , that means the full size of the rom is accessible by the computer .

Could be that this is a byte/swap or setpatch problem .

Can you drop a copy of your setpatched rom on here for me to look at ?

What are you using to burn the rom ?