.
 
 
Model 5580
 Pictures for Processor complexes stored in this page are originally created by 
     Louis  Frederick Ohland.   You can see original pictures for  PS/2 complexes 
at   Louis's  complex page



5580-Y

 General Information
 Y Complex
 W complex
 DX2/66 Up Grade ( Type A )
 SCSI Limitation

 Complex Matrix PS/2 vs PS/55
 Model 8595/9585/9595
 Some tips for 5580/5560
 Clock Lodding

General information
Model 5580 is almost same with PS/2 model 9595 which uses 1 serial 1 parallel planar. 
Basically there are 2 variations of this model.
One is  -Yxx model which uses PS/2 type-K -ish complex and the other is -Wxx model which uses PS/2 
type H/L -ish complex.  ( both have different BIOS P/N  from whose used on PS/2 complexes )
As a matter of fact 580 doesn't support Syncro Streaming Data Transfer mode.
Entry server models, they were.
 

Planar
Uses 1S1P planar. I don't see any distinctive difference from Mod. 95 1S1P planar ( except BIOS EPROM ). 
Click here for 1S1P planar outlined by Louis.

80-Yxx  Processor complex  P/N 79F2458 5580_Y complex
Base clock is  33MHz and a 486DX-33 is loaded on the socket. 
A daughter board for 256K cache  is available as an option (  P/N  79F1294  ECC  63993A  ). 
Memory count is.... very slow like 5560-W0x complex. 

         * Original image was outlined by Louis F. Ohland * 
 
 

As you can see,  80-Y complex doesn't have overdrive socket. 
The complex does not have solder pads just like as  Type-1 DX 2-66 upgrade,  but 80-y complex uses EPROM for BIOS while  PS/2 type-1 DX2-66 upgrade ( 92F0145 ) uses EEPROM ( Flash ROM ). 
The Board P/N is different from 61-W0 complex but all the parts and chips used are same except  OSCs.   Same BIOS ( P/N 79F0981 ) is commonly used for 80-Y and 61-W0 . 

CPU upgrade
Easy to use DX2-66.  You can also use DX4-100, am5x86-133 and Cyrix5x86-100 together with 
an interposer. 


 

80 -W0x/W2x  P/N 07G3153
Processor complex is completely same with the one used for 5560-W1/W2 model. 
Requires REF/DIAG specified for 5580-W ( those for 5560-W1/W2 could not be used fro 5580- W/W2 ). 

    Complex information
    P/N    07G3153
    REA  19602276
    BIOS 07G0463 ( IBM 1981- 1991 )
    Silk print on lower left corner shows " IBM FRU 95F4299",  "ANOIBM486SXB25". 

    CPU 
    5580-W0x   i486SX25 sits on socket position U8 of 80-W0 complex.
    5580-W2x   Earlier models shipped with i486SX25 + ODPDX2-50 ( ID#06G7015 ) on the socket U9.
                        Later models shipped with i486DX2-50 alone on the socket U9.
 

    2002.12.22
    I got a following message while I was creating reference diskette for this model.
    Seems  "Canberra" is the code name for 5580-W .  Does "P23T" mean DX4ODP ? 
 

DX2-66 Up grade Type-A   P/N 66G1987 
Memory count is extremely fast. 
Supports Syncro Streaming data transfer feature and 2.88MB floppy. 

                * Original type2-H GIF image was outlined by Louis F. Ohland * 
 
U1    74F32
U2    74F125D
U3    74F00 
U4    74F04
U6    74F760D
U10,U14-U18   74F244
Y1                  66.666MHz
U5/PA2          59G1883 
U7/DMA32    10G7808 
U9/PGA         DX2-66
U12/PF2       59G1885 
U13/BIOS      P/N 66G1095 



SCSI limitation
5580-Y/W models use older type of  long SCSI /A with cache ( SPOCK  6451018 ). 
Even though  a newest SPOCK  is used , 5580 can't handle a hard drive larger than 1G as an IML drive. 
In order to get full potentiality of newest SPOCK, 8595-J type-3 complex or any other type-4 complex should be used together. 

Can 5580 take pure PS/2 complexes ?
The answer is  "Yes",  using correct reference and diag disks for the PS/2 platform you  want to use.  Some of my friends succeeded to configure a Type3 DX50 complex and Type4-N/Y complex on 5580. 
No reports regarding PS/2 Type 1 or Type 2. 

Comparing chart for processor complex.
Following chart is originally explained by Mr. NATICA ( pseudonym ), one of true MCA enthusiasts in  Japan who helped me in many ways to play with my PS/55s. 
 
PS/2
CPU
PS/55
P/N for PS/55
O/D socket
Remarks
 Type 1-G  486SX20/25        --------       -----    ------  
 Type 1-J  486DX25  5561-W0x  79F0087  None  
 Type 1-K  486DX33  5580-Yxx  79F2458  None  
 Type 1 U/G   486DX50      -------     ------    -----  
 Type 1 U/G  486DX2-66      -------     ------    -----  
           
 Type 2-H  486SX25  5561-W1  07G3153  Yes  
 Type 2-H  486SX25  5580-W0x  07G3153  Yes  
 Type 2-H  486DX2-50  5561-W2x  07G3153  DX2 loaded  
 Type 2-H  486DX2-50  5580-W2  07G3153  DX2 loaded  
 Type 2-L  486DX2-50      -------     ------    -----  
   --------  486DX2-66  Opt for 5580   66G1087  Solder Pads *1, *3
   --------  486DX2-66  5560-N  35G2951  Solder Pads *2, *3

 

Type 3 and all of Type 4 Complexes are commonly used for PS/55 8595-J and Mod. 95. 

I guess PS/55 66G1087/32G2951 are both Japanese unique complexes. No equivalents  in PS/2
can be observed in Louis's page or in MCA Adapter Picture page maintained by Dennis and Peter. 

Remarks
  *1 This complex is called as  "DX2-66 Up Grade Type A". 
  *2 This complex is called as  "DX2-66 Up Grade Type B".  Originally equipped in 60-Nxx. 
  *3 Base clock is 33MHz but is apparently different from  "DX2-66 upgrade 92F0145". 
       Both Type A and B use big  DIP type EPROM. 

Type-A  can be used both in 5560 and 5580 while Type B can only be used in 5561. 



Some tips for 5580/5560
 1. Ref  Ver. 1.21 is what you should have. 
     If you have  more than 16MB of RAM in total,  you need a reference diskette Ver.1.21 to let the system recognize
     whole RAM installed.   ( Ver. 1.11 for 5561 ) 

 2. Under Win95 the system report only 16MB of RAM though you have more than 16MB of  RAM. 
       Replace HIMEM.SYS of Win95 to the one in PCDOS J7.0 then Win9x will see actual RAM you installed.
       This symptom is commonly observed on the other 486 IML machines but  not on EIDE machines such as 5551-R/L
       and 5530-L. 

 3. F/W SCSI-2/A can't be used alone in 5580 with original complex. 
      Not sure about this issue.   It is told that PS/55 BIOS doesn't contain SCSI code (maybe). 

      I haven't yet tested but it is told that ;
      If you want to use F/W SCSI-2/A on 5580, you need to use an usual SPOCK SCSI /A  together with  F/W SCSI-2/A. 



Server models other than 5580.
      9585, 8595-J with type- 3 M platform and  9595 N/P/Q  are same with corresponding PS/2s. 
All models were originally equipped with Japanese ref/diags,  but they can accept English Ref/Diags too. 
 Japanese models of 9595 are all 2S2P type with Type 4 platform.  No differences can be observed except  model-type code and  PS/55 badge. 

I have nothing to say about these machines. 
You can find a lot of info at  Ardent Tools of  Capitalism  ( mirror page) created by Louis Ohland. 
The earlier version is available here.



Clock Rodding for PS/55s.
   Clock rodding is, err.., evil way especially for PS/2 and PS/55.   What we expect for these machine are  stable operation and reliability.  So it is not recommended to change OSC for 20 - 28MHz inner speed.
International Business Machines do not expect clock modification.  I was just a MAD.   It's better to keep your MCA box *clean*.   But anyway here we go..... 

   Japanese Type-1 J with 25MHz OSC for 61-W0 can take 30MHz OSC without any trouble. But all other complexes couldn't take any higher OSC in stead of originally equipped OSCs ( actually  speaking,  one can  use 35MHz OSC in stead of  33MHZ for some complexes ).    In most cases, you  need to modify BIOS for changing base clock of PS/55. Without BIOS  modification,  system will hung up during  POST with 102 error code. 
    I didn't ( well actually "I still don't " ) have any knowledge  to read or write  BIOS.  Mr. Sugiura, one of my friends who knows about reading /writing BIOS,  examined BIOS of 5561-W1x complex and he MODIFIED  the BIOS so that 61-W could boot without  102 error.   According to him, there are several  lines indicating to jump up to 102 error section during POST routine.  He couldn't determine the line which exactly related with changing of the system clock.  Then he rewrote all concerning lines not jumping to 102 error section.  I don't know if his modification includes some bugs or not. But, so far, BIOS ROM with his modification works fine at least in my systems.  He kindly examined another models'  BIOS ROM and burnt new EPROMs for me and for other members of the forum. 

 List of models and complexes modified by Mr. Sugiura.
 Model    Original OSC      Replaced to 
 Up Rated  from/to
   Type of BIOS ROM
 5551-N     66.66MHz
 84MHz
 33 to 42MHz
 PROM, write one time 
 5551-W0     40.00MHz
 60MHz
 20 to 30MHz
 PROM, write one time 
 5561-W1/W2     50.00MHz
 62MHz
 25 to 32MHz 
EPROM
 5580-Y     33.33MHz
 40MHz
 33 to 40MHz
 EPROM
 Type-A upgrade     66.66MHz
 80MHz
 33 to 40MHz
 EPROM
 Type-B upgrade     66.66MHz
 80MHz
 33 to 40MHz
 EPROM
 modification story is available in this page too.
 Line #3 amended from 5551-W1/W2 to  5561-W1/W2 on 10.11.2002 
 
 

PS/55 Index