PDA

View Full Version : IBM 5162 XT-286 IDE support



Chromedome45
May 20th, 2011, 06:43 PM
Hi all, I picked up a 5162 motherboard from codemicro at a decent price and was wondering what does one have to do to get an IDE drive working in this thing? None of the predefined drive types match and I have no way of manually entering CHS. Any suggestions I'm lost...:confused::( I even tried the XT-IDE and no go there either.

mikey99
May 20th, 2011, 07:17 PM
It would be the same as using IDE in a 5170 AT, any 16 bit IDE controller with it's own BIOS should work okay.
For the predefined drive settings , I think it probably doesn't matter what you set it to. First try setting it to
no hard drives and see if it works.

However..... I would think the XT-IDE should work okay too ......
Does the XT-IDE BIOS load and do you get to the boot menu ?

Chromedome45
May 21st, 2011, 06:03 AM
On the XT-IDE I never even see the BIOS screen come up. But then I may need to reflash to the AT version.

per
May 21st, 2011, 06:55 AM
The XT-286 expects the later 5170-style disk controller, and it will give errors if you try to replace it with alternative controllers.

Chromedome45
May 21st, 2011, 07:26 AM
So I am begining to find out. Do yoy know the Part number of the board? IBM's FRU as they call it? Also I would want a low profile board that would fit in an XT Case. Most AT style boards as you know are a little taller. Maybe you can look at your controller there Per.

per
May 21st, 2011, 07:29 AM
http://oldibmpc.sitesled.com/Expansion%20cards/IBM%20Corp.html#17

This is the exact controller which came with my XT-286. The card itself is actually made by Western Digital, but sold by IBM as an OEM product. The WD model number is WD1003-GRY.

Chromedome45
May 21st, 2011, 09:30 AM
So basically without this WD/IBM OEM controller I'm kinda out of luck huh? Any other suggestions so fra even my ADP-50 IDE isn't woking either. :(

per
May 21st, 2011, 09:45 AM
So basically without this WD/IBM OEM controller I'm kinda out of luck huh? Any other suggestions so fra even my ADP-50 IDE isn't woking either. :(

You can of course replace the BIOS with a third-party one, but that would be slightly controversial depending on to what degree you consider originality.

I don't remember if it just report the error and continues, or if it reports the error and halts.

Chuck(G)
May 21st, 2011, 10:26 AM
It's not that hard to find an older IDE controller--when EIDE came out, there were several third-party boards with their own BIOS that could support them. The Promise EIDEMax is one, for example (http://www.artofhacking.com/th99/c/P-R/20785.htm).

Chromedome45
May 21st, 2011, 10:44 AM
Well good news I got it working my old Oak CGA/EGA/VGA combo card was knocking out my IDE conrollers (address conflict?). I now have in there temporarily the IBM MDA and my ADP50 is working fine. HD access etc is all good. For some reasons my one combo card that does have a BIOS gives me a 601 error. So I used another combo card and it seems to work okay. Both of these boards are made by SIIG. So still doing some minor tweaking to it. :D

per
May 21st, 2011, 11:29 AM
The 601 error is the BIOS which can't find the IBM disk controller, but this really doesn't matter now as the adapter you use replaces the IBM BIOS fixed-disk routines.

If you want to check out exactly what the POST looks for regarding the original disk controller, you should consult the BIOS listing in XT-286 technical reference:
http://filedump.glitchwrks.com/manuals/IBM/techref/5162/

modem7
May 21st, 2011, 03:03 PM
If you want to check out exactly what the POST looks for regarding the original disk controller, you should consult the BIOS listing in XT-286 technical reference:
http://filedump.glitchwrks.com/manuals/IBM/techref/5162/
Per. Thanks for scanning your Technical Reference. I know that many people will find it useful.

Chromedome45
May 22nd, 2011, 09:46 AM
Just 1 more question then. Is it possible to use the AT BIOS in this board? I have seen updated BIOS available for the AT computer.

per
May 22nd, 2011, 09:57 AM
Just 1 more question then. Is it possible to use the AT BIOS in this board? I have seen updated BIOS available for the AT computer.

Yes, but as I mentioned; it breaks the originality (in the case of perfection). Whatever you do, I suggest that you don't throw away the original BIOS if you replace it.

The most signifficant difference between the AT and XT-286 is the refresh logic. The RAM in the XT-286 somehow needs to be refreshed more than in a regular IBM AT according to the techref, and the only way to see if there is issues with another BIOS is to try the other BIOS.

Chromedome45
May 22nd, 2011, 11:06 AM
Ok well I tried to use the AMI BIOS that was on that site in Australia and the system would not post. Just sat there. Tried moving the EEproms between the 2 sockets and nothing happened. I have not tried an actual AT BIOS and don't intend to.

per
May 22nd, 2011, 12:14 PM
Ok well I tried to use the AMI BIOS that was on that site in Australia and the system would not post. Just sat there. Tried moving the EEproms between the 2 sockets and nothing happened. I have not tried an actual AT BIOS and don't intend to.

It may be due to the use of EEPROMs rather than EPROMs. Does the EEPROMs you use have software dataprotection enabled? Also; did you correctly insert them and checked that everything was OK?

Chromedome45
May 23rd, 2011, 04:26 AM
The EEPROMs are both ATMEL so I think they have software protection. I did insert them correctly but not sure which one goes into which socket.

per
May 23rd, 2011, 06:05 AM
The EEPROMs are both ATMEL so I think they have software protection. I did insert them correctly but not sure which one goes into which socket.

I just compared the pinout of the 27256 (EPROM) and the 28256 (EEPROM), and they are actually not pin-compatible. The signifficant difference is that A14 is on pin 1 of the EEPROM (pin 27 is /WE), while it is on pin 27 on the EPROM (pin 1 is VPP).

The regular IBM 5160 XT supports both as it routes A14 to both pin 1 and pin 27, but the XT-286 does not.

The reason of this incompability is problably due to parallel development. The 2864 was based on the 2764, which is fine for compability, but the 28256 is an evolution of the 2864 rather than being based on the 27256 (which is an evolution of the 2764).