HP Forums
HP-42S Saturn Address lines? - Printable Version

+- HP Forums (https://archived.hpcalc.org/museumforum)
+-- Forum: HP Museum Forums (https://archived.hpcalc.org/museumforum/forum-1.html)
+--- Forum: Old HP Forum Archives (https://archived.hpcalc.org/museumforum/forum-2.html)
+--- Thread: HP-42S Saturn Address lines? (/thread-60881.html)



HP-42S Saturn Address lines? - Nelson M. Sicuro (Brazil) - 07-20-2004

Hi to all,

I'm trying to discover if the Saturn CPU from the HP-42S have more than the A0..A15 address lines.

I think that some non-connected CPU pins that ends on test-pads on the PCB following the sequence of the address lines (after A15) can be used as address lines for the main RAM - *IF* those pins are indeed address lines!! I don't know very deeply the Saturn processor and I don't know if its address mode is linear on the 0.5MB address area (1Mega-nibble). There is the chip-select (CS) issue also, I think that the ROM-CS and RAM-CS are hardwired on some address window. Any information on this?

I don't have a logic analyser, but I'll try to borrow a oscilloscope to peek around. If these lines can be used for addressing the RAM, the extra RAM can easily be added - 128KB maybe???

Anyway, it is easy to put 64KB of RAM in the space between $20000~$3FFFF to do some extension of its functions/OS, and if at least I can use the A15 on the normal RAM space I can put 64KB there too (128KB total).

Best regards,

Nelson


Re: HP-42S Saturn Address lines? - Christoph Giesselink - 07-20-2004

> I'm trying to discover if the Saturn CPU from the HP-42S have more than the A0..A15 address lines.

The 1LR2 Lewis chips has 17 (MA[9:0] shared with OR[0:9] and MA[10:16] shared with IR[0:6]) address lines for byte access.

> I think that some non-connected CPU pins that ends on test-pads on the PCB following the sequence of the address lines (after A15) can be used as address lines for the main RAM - *IF* those pins are indeed address lines!! I don't know very deeply the Saturn processor and I don't know if its address mode is linear on the 0.5MB address area (1Mega-nibble). There is the chip-select (CS) issue also, I think that the ROM-CS and RAM-CS are hardwired on some address window. Any information on this?

The MMU for the two CS pins for external ROM and external RAM access is programmed at manufacturing process and can't be changed later. The address areas are $20000-$3FFFF and $50000-$5FFFF.

> I don't have a logic analyser, but I'll try to borrow a oscilloscope to peek around. If these lines can be used for addressing the RAM, the extra RAM can easily be added - 128KB maybe???

> Anyway, it is easy to put 64KB of RAM in the space between $20000~$3FFFF to do some extension of its functions/OS, and if at least I can use the A15 on the normal RAM space I can put 64KB there too (128KB total).

Both not possible because of the fixed programmed MMU with 64KB at $20000-$3FFFF and 32KB at $50000-$5FFFF. You will not get any external CS signals outside these memory areas.

The additional pins on the test-pads maybe the 3 MODE[0:2] pins and/or the Saturn bus pins BUS[0:3], NSTR and NCD.

Best Regards,

Christoph


Edited: 20 July 2004, 6:05 p.m.


HP-42S Saturn Address lines? - V-PN - 07-20-2004

The address space is linear.

'VPN'

Edited: 20 July 2004, 5:55 p.m.


Re: HP-42S Saturn Address lines? - Nelson M. Sicuro (Brazil) - 07-21-2004

Hmmm.... My only solution for having mode RAM/ROM (FLASH?) on the 42S is bank switching by some glue logic - I have an idea to put another microprocessor inside it to handle communications (very low power MSP430F1121 line) and maybe it can control the RAM/FLASH CS as well. It is just an speculation as I don't know if it can handle the address/data bus of the Saturn fast enough with its (MSP430) 8MHz.

Anyway, having those 64KB to play with is a good challenge!

Best regards,

Nelson