The Sinclair QL Forum

t0nyt wrote: Fri Feb 21, 2025 9:10 am From the point of view of me getting my 0k board working I’ve come to the end of the journey for the time being as I’ve realised that after trying the PAL from the memory board the 0k isn’t even recognised as a floppy drive any more

Thought I'd give it one more go, so I re-seated some chips and it's working again as a floppy controller!

I'll try soldering up the 3 missing legs in the next few days

Thanks

Hi Tony

Bear in mind those other slight cct differences that were mentioned before, likely as not passive parts needed to disable the DSMC signal during memory access to the RAM area between 256-512K and/or refresh delay timing.

The PAL equations are likely very similar to other memory expansions (address decode, DSMC disable and DRAM refresh cycle timing) - There's only so many ways to design such a cct, so you're probably very close to getting the 0k board to work with the added DRAM, but of course, it's a matter of effort vs perceived benefit

Anyway, good luck.

martyn_hill wrote: Fri Feb 21, 2025 10:18 am Hi Tony

Bear in mind those other slight cct differences that were mentioned before, likely as not passive parts needed to disable the DSMC signal during memory access to the RAM area between 256-512K and/or refresh delay timing.

The PAL equations are likely very similar to other memory expansions (address decode, DSMC disable and DRAM refresh cycle timing) - There's only so many ways to design such a cct, so you're probably very close to getting the 0k board to work with the added DRAM, but of course, it's a matter of effort vs perceived benefit

Anyway, good luck.

Thanks Martyn

"it's a matter of effort vs perceived benefit" is a major part of my thinking now

I know the 0k board works as 256k just by putting the PAL from the 256k board into it, so feel fairly comfortable the PAL is the only issue

Given these boards are only for use when testing/playing with various things (I use an SGC in my main QL and normally a Trump 768K in this QL) perhaps I should just settle on having a 256k and a 0k in my collection rather than risking having to bin the 0k (weirdly it doesn't get on well with other memory boards though, so has very limited use)

Thanks
Tony

Hi,

As stated earlier, I have a PCML board, which the RAM expansion does not work, so in the absence of a copy of the working PAL code, I will try and work out how to get the RAM expansion working.

As Tony, says, it will be good to use it with a QIMSI, which I use a Tetroid TDI interface, with the CF Interface disabled, to allow the QIMSI to work.

Jackpot!

I managed to solder 3 new legs on the PAL chip and now Minerva reports a total of 384k of memory

Derek, if your equipment can read a 16R4 PAL, and you have the time, I'd be willing to risk posting the restored PAL as it's a spare board

Can't guarantee the legs won't fall off when it's extracted from the socket though!

Thanks
Tony

t0nyt wrote: Fri Feb 21, 2025 1:42 pm Derek, if your equipment can read a 16R4 PAL, and you have the time, I'd be willing to risk posting the restored PAL as it's a spare board

Can't guarantee the legs won't fall off when it's extracted from the socket though!

Thanks
Tony

Hi Tony,

Is the PAL chip type: DMPAL16R4ANC ?

My programmer supports many PALR4 chips, so I would say it will read the PAL if the security fuses are not blown to prevent copying.

I will send you my address by PM

Derek_Stewart wrote: Fri Feb 21, 2025 2:07 pm Is the PAL chip type: DMPAL16R4ANC ?

Yes, that's the correct chip

Hi Tony,

I have 2 PCML boards, one complete and the other needs a ROM, FDC, PAL, also the oscillator at the FLP connector is broken off.

The complete board which the PAL chip doesn't recognise the RAM, Is a PAL16R4ACN

But I would think any PAL16R4 chip would do.

Have managed to extract the PAL with the new legs still attached

Will post tomorrow

Thanks
Tony