After finding a couple of IDE to SCSI bridges, I had to go to way back machine to find information so I am reposting it at the end.
I can confirm the last three dip switches do set the scsi drive number with either bridge. On the ide to SCSI bridge from a Power Book duo (pictured with the Lattice chip above the switches/kaplan tape)the adapter works with the apple labeled IDE drive and other (tested IBM and Fujitsu less than 2 gb) drives.
The other pictured bridge (without the lattice chip above the dip switches), even set with dip switches 3 and 4 on (or off for that matter), does not terminate in a powerbook 160....as soon as I plug in an external terminated Zip drive, only then does boot from the internal drive continue.
The setting of the first 3 dip switches does not alter the function, as far as I can detect, of a functioning boot up IDE drive plugged into the bridge. (I am talking left to right in the photos...watch out, because after looking at the photos, I can read the dip switch numbers)
REPOST:
There are two known versions for the board (chips listed left to right topdown):
- symbios logic 53CF96-2 (narrow SCSI HBA);
- a pair of Toshiba TC558128AJ-15 (15ns 128Kx8 CMOS static RAM);
- a pair of ALS245A (buffers);
- Lattice pLSI1016 (PLCC);
- LSI KL5C80A16CF (Z80 instruction compatible microcontroller);
- Atmel AT27C256R-70RC (70ns 32Kx8 CMOS EPROM).
- symbios logic 53CF96-2 (narrow SCSI HBA);
- a pair of SEC KM681001BJ-15 (15ns 128Kx8 CMOS static RAM);
- a pair of ALS245A (buffers);
- LSI KL5C80A16CF (Z80 instruction compatible microcontroller);
- Atmel AT27C256R-70RC (70ns 32Kx8 CMOS EPROM);
- Lattice pLSI1016 (PLCC).
The DIP switches
Left to right:
0
dunno
1
dunno
2
dunno
3
SCSI bus termination
4
SCSI bus termination
5
SCSI target bit 2
6
SCSI target bit 1
7
SCSI target bit 0
IDE drive geometry
as far as i can tell the board only understands drives with maximum of 15 heads and 63 sectors per track. the cylinder number is limited to 16384. the controller does not honour the total number of sectors on the drive that limits the total capacity to a little bit less than 8G.
drives with more than 15 heads seem to allow the device to probe but fail to accomplish any actual data transfers resulting in controller timeouts. that sounds like an overflow and thus division by zero on the board's controller (;