Olivetti M240 Repairing

[demonlg]

Hi to all from center Italy, and sorry for my english,

i have necessity of support for my two Olivetti M240, both have same problem but i'm stuck on this from 2 week without solution. On line i d'ont found any schematics for testing with scope or other instrument.

Pratically i'm a computer collector and 10 year ago i have bought two M240, first one not working and second one yes, with hdd all perfect, i have installed (10 year ago) an isa VGA card 16 bit inserted in a 8 bit slot and it's woriking perfect.

Two week ago i have turned on this computer but i dont' have video output, not boot beep, nothing! I have tested VGA card in other ISA computer and it's working, i have reinstalled original EGA card but nothing.

I have checked jumper settings but nothing. Then i have replaced mainboard with the other one not working and i have same problem. Then i have replaced ROM from first board to second board and in any case i have a boot beep but nothing on video, but this beep is randomly.
The leds on keyboard (at this moment is a keyboard 1 model) stay on fixed when the beep is not present, and when i have a beep the led flash fast continuosly.

I have replaced bios battery with new one.
I have setted jumper in 256k mode ( i think so the first ram bank is composed of four 41464 chip (64k X 4 =256k) but same problem, nothing.
I have replaced all chip socketed (floppy controller serial controller and PAL chips) and CPU tested on other 8086 machine IBM PS2 8530, CPU are ok other chip i dont' know but nothing.

I don't know what is the problem, i think so is a RAM problem at first bank 256k, or any logic chip on the mainboard, but without schematics it's very difficult to repair it for me.

Any help is appreciated.

Thanks

Emanuel

 

[modem7]

I have setted jumper in 256k mode ( i think so the first ram bank is composed of four 41464 chip (64k X 4 =256k) but same problem, nothing.

41464 = 65,536 x 4-bit
Therefore, two 41464 (two of 4-bit) are needed to get a byte (8 bits).
Two 41464 = 65,536 x 8-bit = 65,536 x byte = 64 KB.
Four 41464 = 128 KB.

However, the M240 video at [here], at about the 9 minute mark, indicates 640 KB on the motherboard using twenty 41256 chips (20 x 262,144 x 1-bit = 20 x 32 KB = 640 KB).
I.e. The M240 is designed to use 41256 chips, not 41464.

 

[demonlg]

Hi modem7,

yes but my two boards have 20 chip of 41256 and at the right of this i have 4 chips 41464, from the 20 chips and the top of System Roms.
Then another question if possible, anyone of you have the model type or quivalent code for Flyback transforem of the M24/M240 12" monochrome monitor DSM2412?

I have two of this monitor, one working perfect and one have a flyback blowned, but the two monitor have different internal board, one broken is market Hantarex, the other one i don't know.

 

[modem7]

yes but my two boards have 20 chip of 41256 and at the right of this i have 4 chips 41464, from the 20 chips and the top of System Roms.

That makes better sense; the motherboard RAM includes parity bits.

I see online that the switch settings for RAM are: {none}{256 KB}{512 KB}{640 KB}

Looking at how {20 of 41256, plus 4 of 41464} and those switch settings all work together, I come up with the following bank breakdown.
I have not been able to find anything online to verify that breakdown.

+--------+---------------+----------------+-----------+ | Bank | | | Bank size | | number | Data bits (8) | Parity bit (1) | (data) | +--------+---------------+----------------+-----------+ | 0 8 x 41256 1 x 41256 256 KB | +--------+---------------+----------------+-----------+ | 1 8 x 41256 1 x 41256 256 KB | +--------+---------------+----------------+-----------+ | 2 2 x 41464 1 x 41256 64 KB | +--------+---------------+----------------+-----------+ | 3 2 x 41464 1 x 41256 64 KB | +--------+---------------+----------------+-----------+ 640 KB total

 

[demonlg]

I will try to desolder the four 41464 chip and test it separately, i don't have a static ram tester but i have to test the chip on my C64C with socketed same RAM (two in this case for 64k). Then if this four chip are working i will try to desolder all 20 chips 41256.

I think so the problem on my two board is on any RAM bank, because all other socketed chip are tested and working. If the problem is on the logic chip i have no chance to repair it, because i don't have schematics to check.

Tomorrow i have tested the reset RST on CPU and the CLK timing with my scope, in the CLK pin 19 of 8086 CPU i have one perfect sinusoidal signal on both boards, the same of my working M24 board, then on the RST line i have 0 volt for 2-3 second then 5 volt, the RST line is working.

Then i have tested all 5v on ol chip on the board and all is fine 5,01 volt on all of the VCC pins of chips.

Thanks for the moment, i upgrade post with my test.

Emanuel

 

[TiBos Retro Computers]

41464 = 65,536 x 4-bit
Therefore, two 41464 (two of 4-bit) are needed to get a byte (8 bits).
Two 41464 = 65,536 x 8-bit = 65,536 x byte = 64 KB.
Four 41464 = 128 KB.

However, the M240 video at [here], at about the 9 minute mark, indicates 640 KB on the motherboard using twenty 41256 chips (20 x 262,144 x 1-bit = 20 x 32 KB = 640 KB).
I.e. The M240 is designed to use 41256 chips, not 41464.

Here is a Close up of my M240