Hard disk Driver Card WD1002A-WX1 Without Bios

[MauriceH]

Have a question.

The WD1002A-WX1 has NO Bios mounted.
It is never been with a Bios. See picture.

W3 is open, so Bios is Disabled.No jumper can be installed also.
W7 is hard wired and so IRQ=5

How can I configure it for a XT machine?
G=C800:5 does not work at debug mode.
Ofcourse its missing a Bios.

Put it in a 386 machine and HDD Hardware failure
With Biossetting drive 0 = Type 2
This 386 has a 8-Bit slot.

To test this set-up I did same with a WD1002S-WX2 card and his working ST-225.
Pull W3 to setting Disable Bios, and in bios 386 the type 2 (ST-225 setting)
Again HDD Failure, And that set-up Card and drive work in a XT machine.

How is it with this WD1002A-WX1 card with no bios.
Where there XT's with its own Harddisk Bios?

 

[modem7]

Put it in a 386 machine and HDD Hardware failure
With Biossetting drive 0 = Type 2

Because you set a type number into the AT's BIOS/CMOS SETUP, the POST went looking for an AT-class MFM controller and drive.

XT-class MFM controllers and AT-class MFM controllers are different. Refer to the diagram at [here].

For an XT-class MFM controller fitted into an AT-class computer, the AT's BIOS/CMOS SETUP would be set so that there is no hard drive (ie. type 0 = there is no AT-class MFM controller and drive that the AT motherboard's BIOS needs to 'manage'). The BIOS on the XT-class MFM controller would be what 'manages' the XT-class MFM controller and drive.

An issue is compatibility (e.g. bus timing), but the Western Digital information at [here] indicates 'AT' compatibility. Supported by the 'XT mode/AT mode' jumper shown at [here]. It is not indicated what the 'XT mode/AT mode' jumper does. Maybe in AT mode, the jumper just puts the controller into 'latch' operation.

You need to find an appropriate BIOS for the card.

 

[modem7]

How is it with this WD1002A-WX1 card with no bios.
Where there XT's with its own Harddisk Bios?

I browsed through my records.

There is an email trail from a member here.
He had an Olivetti M24, and the M24 contained a DTC 5150BX4 that was missing its BIOS ROM chip.
After research, he later let me know that on the M24's motherboard, switch 3 on switch bank DIPSW-1 disables/enables an onboard HDD BIOS.

 

[modem7]

Member SomeGuy made a copy of the BIOS ROM on his WD1002A-WX1.
That copy is at [here].
It is an 'F300' version, which means that it has 'dynamic' functionality - see [here].

If you are unable to program a 2764 type EPROM yourself, the people/companies at [here] can provide the service.

 

[MauriceH]

Thanks for info and advise.

Modem7
I put in S1-8 to AT mode, Because there is No Bios original mounted I use the AT-Bios for type setting.
Also don't get it why the settings S1 1-4 will be needed to identify hard disk settings But used the ST-225 setting any way.

I could indeed put a Eprom in it, no problem. 2764 I have (Don't forget the W3 jumper )
That way I can test if the card at least work in a XT.

Think It came out of my Unisys 3137 .That uses a Bios setup Disk, I still have that disk, not the pc.
Just found out that the machine was a 286 pc.
I'll have to dig out my WANG 286 PC to try if it will work on a 286.

AND Modem7 Thanks for the Bios WD1002S-WX2, Have that Card and That also missed the BIOS.
That card had already the IC-holder so fitting was no issue.
Burn that Bios you put on, and Its working perfect.

Now reading the Book: CSC Hard Drive Bible - Edition 8 - 1996.pdf

I'll be back.

 

[modem7]

I put in S1-8 to AT mode, Because there is No Bios original mounted I use the AT-Bios for type setting.

Setting the XT/AT jumper to 'AT' does not turn the controller into an AT-class one. We know that because it is not possible for an 8-bit card to generate interrupt 14.

The information at [here] for a different XT-class Western Digital controller, the WD1002-27X, contains "XT Buss Mode" and "AT Buss Mode", so in AT mode, the WD1002-27X must be adjusting things to make the controller better suited to the ISA bus on an AT. The XT/AT jumper on the WD1002A-WX1 is probably the same.

And in another set of WD1002-27X information, at [here], there is some text about operating the WD1002-27X controller (XT-class) in an AT:
"Drives operating on the WD1002-27X are not programmed into the AT's configuration memory. Therefore, configure the IBM Advanced
Diagnostic program, Setup, for no(zero) drives and set the controller jumpers as the primary controller (W4:2-3; W8:2-3)."

Also don't get it why the settings S1 1-4 will be needed to identify hard disk settings

When an XT-class hard drive controller is operating in an AT-class computer, the code that controls/manages the attached drive is in the BIOS ROM of the XT-class hard drive controller. That code does not look at the AT's CMOS/BIOS SETUP. I doubt that the code even knows that it is running in an AT-class computer.

 

[MauriceH]

Get the WD1002A-WX1 original without Bios, Running.

Put a IC Holder on the PCB and a W3 Jumper.
First as suggested I burned a 2764 with the Super Bios V2.4.
The card did a run on the Bios but does not activate the drive it self.
It hangs on Low-Level-Format and No LED Light or activate of stepper motor.

EDIT:Just realised W5-JUMPER, Bios 32k or 64K ! But that did not solve the problem.

First thought the card was really defect, and does not signal to the drive.

As I have a second WD1002A-WX1 with Bios, a 2732A Rev 6 , I did some
measurements on Resistor to some components to see if I could find some differences.
So Found also the cards are not quit the same! Remark f.i. a VALOR DL2879 block.Could not find any thing on that part.
Then Noticed a difference in Resistor measurement on C26 273 Ohm on the "defective no bios card" and 876 Ohm on the other.

Then I decided to read in the other BIOS and burn a new 2732A to put on the defective card.
As I don't want to mess up a good working card.
Well that was the solution!

Card is now perfectly working.
Conclusion that Super Bios is not quite useful in this WD1002A-WX1.
I found several differences on the internet with several combinations Bios and VALOR DL2879, also some components as R & C's.

So again as this is working in a XT machine, now the AT 386.
In Bios 386 disabled all hard drives, Now the WD1002A-WX1 is been recognised and BOOTs the system.
Jumper S1-8 is NOT set! No AT-mode been used.
If I set S1-8, card to AT-Mode again no Hard Disk, even the right Type setting in 386 Bios does NOT work.

Hope to others that this will give options on how to get those old MFM-card going.

I have again a presumed defective MFM-card, up and running.
Its all about settings, jumpers, Bios etc.
Pictures are still with that Super Bios, But now the REV 6.00 1985 is put on.

 

[MauriceH]

WD1002A-WX1 Bios Rev 6.00 1985

WD1002A-WX1 Bios Rev 6.00 1985

@ Modem7

As I did a read of my Bios WD1002A-WX1, I did an upload of this file.
Rev 6.00 1985 Label is "10/20/30 "
Feel free to use it on your site, and distribute to others.
It is a EPROM 2732A Situated to the card edge.
Pins 3-14/15-26
Attached inside Zip file also a TXT file with information.

Maurice

View attachment BIOS ROM on WD1002A-WX1_REV-6_1985_Type2732A.zip

 

[modem7]

As I did a read of my Bios WD1002A-WX1, I did an upload of this file.
Rev 6.00 1985 Label is "10/20/30 "
Feel free to use it on your site, and distribute to others.

The revision number of BIOS ROM's that Western Digital use on their XT-class and AT-class hard disk controllers, is a long number, like 62-000042-11 and 62-000215-060 and 62-000089-030. For example, in the OEM manual for the WD1002S-WX2 is, "with 62-000042-12 WD BIOS."

In the code of your BIOS is "WX2 Format Revision 6.0". So, 6.0 will be the revision of the format code only. The Western Digital BIOS ROM that has '62-000042-11' on its sticker is a different ROM to yours (different code), and it also has "WX2 Format Revision 6.0" in it.

'10/20/30' is odd. It is not a date, and not even an abbreviated form of a BIOS revision that Western Digital use for their XT-class and AT-class hard disk controllers. The BIOS may be a user-modified version of a Western Digital BIOS ROM, and "10/20/30" indicates to the user what the modification is.

I will add the BIOS to the collection at [here], against WD1002A-WX1, but I will list the revision as unknown.

 

[Malc]

Interesting, Just did a quick google and there seem's to be different variations of the WD1002A-WX1 controller, Component count etc, I have the " F300 " version of the WD1002A-WX1, On the printed label it reads " WD1002A-WX1 F300 J " and the Bios ROM is soldered directly onto the PCB. It looks like the last digits are of the BIOS / Revision that will work with the controller.

 

[modem7]

I have the " F300 " version of the WD1002A-WX1, On the printed label it reads " WD1002A-WX1 F300 J " and the Bios ROM is soldered directly onto the PCB.

That 'F300' is known as a 'WD Feature Number.'

For example, table 2 within the WDXT-GEN user's guide at [here] reveals that there were (at the time of publication) five different feature numbers available for the WDXT-GEN:

F300 = Supports a heads=4/cylinders=615/WPC=307/RWC=none/Step=30us drive, AND extra features such as dynamic operation
F320 = Supports only a heads=4/cylinders=615/WPC=307/RWC=none drive, stepping at 30us
F335 = Supports only a heads=5/cylinders=733/WPC=367/RWC=none drive, stepping at 18us
F336 = Supports only a heads=6/cylinders=640/WPC=210/RWC=210 drive, stepping at 18us
F340 = Supports only a heads=6/cylinders=820/WPC=410/RWC=none drive, stepping at 18us

(F300 being the more desirable due to its optional dynamic operation.)

Looking at the differences, they are things done in code, i.e. 'F300' relates to the BIOS ROM.

 

[Dwight Elvey]

There were a couple things I remember about XT controllers and later AT controllers. I used a XT controller with my NC4000 computer. It is not a x86 processor, it is a stack processor. It would just crash using XT's ROM. So, I wrote my own bios code for the controller.
I also uses a real ST506 drive ( 5Megs ). 'payed $5 or $1 a Meg.
Anyway, back to the controller. The XT controller and the AT controller used completely different register to the bus. The AT used a register set like the original WD1000 series ( notice the last number is a 0 ). The XT used a different register set. This means that you can't send the same instructions to a XT controller that a AT class controller can use.
Both types are easier than the typical floppy controller to write the BIOS for. Both require the typical stuff of the various setups. Things like stepping speed, sector size and such. Both were intended to buffer an entire sector at a time as a minimum. That made writing and reading easy. Fill the buffer and write, empty the buffer on reads.
One other thing, I believe, starting around the ST255 time, you want to use the auto stepping rather than the fixed stepping. If the drives used the fixed stepping, they when really slow.
Any way, there are the two different register sets used and the location of the bits are completely different.
I recall how easy it was to do the hard drive compared to the floppy. The floppy controller did a byte at a time, similar to a serial chip. I couldn't send it to fast or too slow without issues. The NC4000 was so much faster that the controller didn't have time to update the status before it read it after the data write. I needed to add delays.
The NC4000 had no interrupt or dma. I always thought interrupt control on a single process machine was silly. What were you going to do while waiting for the interrupt, especially on reads. The reason you asked for the data is because, you needed it right now and you had to wait anyway.
Dwight

 

[MauriceH]

Dwight,
Thanks for the insight info about registers.

As for This XT WD1002 card that originally came with no BIOS I still can not figure out a machine that will understand that card.

Now I finally have that very old 1987 80286 board running, I thought maybe that will go, but not.
Ok now with the Bios installed it run's, But without a Bios??
Then that jumper AT-Mode, as Modem7 recalled this card is not able to int14, so the use of it, the manual does not
give any clue about running without bios or The AT-mode.

And why having a Jumper with totally no use, That W3 disabling the Bios so CPU won't recognise any hard drive?
You have to unscrew and pull the card, replace jumper, put in ,fiddeling with the screw, so it does not see the drive?
Leave whole card out would be easier.I think.
There must have been MBoards with BIOS that include MFM, I think.

 

[MauriceH]

At Debug That "10/20/30" Bios is "WX2 Format Revision 6.0"
See

 

[Dwight Elvey]

When you have a mother board with a bios, you don't want to install the onboard bios. I can't tell you why your particular system may or may not work with this card. It may be that the card does not work in any system. I do know that if the BIOS you have does not use the right registers it is assured to not work. There are two different register sets typically use in hard drive controllers. The two types of controllers are both MFM on the drive side so MFM or or not it makes no difference. IDE interfaces are completely different. These are not the two controller types I am talking about as the IDE drive actually have the controller on the drive. The card on the buss is just a bus interface.
Most consider a AT to be a IDE type. I can't see how this board has a AT mode.
Dwight

Dwight

 

[gertk]

The card looks strikingly similar to the card I shoehorned into the Kaypro 4-83 (it came from a hardcard with a 20 MB Kalok MFM on it)

 

[Dwight Elvey]

Maybe you could dump the bios and the OP could blow an EPROM and attach it to his board.
Dwight

 

[gertk]

Maybe you could dump the bios and the OP could blow an EPROM and attach it to his board.
Dwight

With some effort (the ROM seems to be soldered into place) that could be possible.
I can not read out the rom from the kaypro as I only connected the IO lines needed to access the controller part of the card.

For the fact why it would ever be necessary to disable the ROM: maybe in a multiple drive setup (more than 2 harddisks) then you would need to add a second card and the roms would conflict?

 

[Dwight Elvey]

It is in the computer as a specific address. You should be able to read it into a file. The ROM should start at C800:0000. You should be able to able to read it with any descent language.
Dwight

 

[gertk]

It is in the computer as a specific address. You should be able to read it into a file. The ROM should start at C800:0000. You should be able to able to read it with any descent language.
Dwight

Alas my Kaypro is not a DOS/80x86 computer

I needed a MFM controller to drive a nice 10MB Seagate full height drive and found two 8 bit ISA hardcards with defective harddrives.
The only controller I could find some info on was the WD controller. The controller is mapped into IO space and no memory space is used.

Wrote a new BIOS for CPM on the Kaypro and finally got it working. It now even boots from the harddisk..