• Please review our updated Terms and Rules here

8-Bit IDE Controller

No XC9572XL. I believe in having margin - especially in old electronics. I just don't believe the MOS 3.3V I/Os will be able to source enough drive current for every odd ball mismatched impedance ISA bus or IDE drive people try and plug in. If a mismatched line causes too much of a draw and the I/O sags even a little, it's possible the drive or bus may not register a one. I know you've done a lot of successful compatibility testing, but I just don't see the point in saving $4.5 when there is an equivalent part in plentiful supply with a built-in LDO and 5V buffers. And nothing on this board will be SMT. Again if both the TQFP-100 and the PLCC-84 packages in plentiful supply, I don't see the choice in the former. This board wont be fulfilled by Seeed but by end users with soldering irons. That's really where the design focus differs. Just my $.02.
 
There's never any offense. :) There are a couple points where the DP design and mine diverged. Was just pointing them out :) I still think there is merit in having two options. I thought the progression was positive on the DP thread until Ian posted the rolled back design without any explanation of why it was reverted to such a old point.
 
Wanted to come back here and make a quick post about a problem I solved on my XT-IDE kit which was from one of the original batch, in case someone has the same issue. Symptom was that the system wouldn't even boot with the XT-IDE card in any slot. Turns out that the culprit was a wonky EEPROM. I have heard mixed reviews on the SEEQ EEPROMS that were included in the kit, mine happened to have something bad on the D6 line (would not write any bit sent along this line). Anyway, got a more reliable brand of 28C64A from Mouser and everything looks A-OK. :)
 
Is there anything new in ISA 8 bit IDE card development ? I would definitily be interested in buying/building one.
 
tingo said:
Is it available with SMD components pre-soldered?
Click to expand...

It really isn't hard to solder SMT packages. Have a look at this page, for example, which solders a 100 pin 0.5mm pitch TQFP. You do need a decent magnifier to check your work, however. An inexpensive triplet jeweler's loupe will do. Dissection microscope is even better.
 
Chuck(G) said:
It really isn't hard to solder SMT packages. Have a look at this page, for example, which solders a 100 pin 0.5mm pitch TQFP. You do need a decent magnifier to check your work, however. An inexpensive triplet jeweler's loupe will do. Dissection microscope is even better.
Click to expand...

That may be, but currently I do not want to expose myself to the level of frustration required for me in order to succeed in that learning experience. YMMV.
 
Tricky at first, yes. Syringe flux makes it a lot easier. Once mastered though it opens up so many parts to DIY use.
 
pearce_jj said:
Feel up to soldering one of these?
Click to expand...

Oohh neat. :) I can handle SMT soldering (have a hot-air rework gun). Has someone already done up boards and are any available for purchase? Looks simple enough where I prolly won't even want a soldermask (unless for some reason I was making a bunch... but doubt it.. if that's the case tho my neighbor has an oven I can use).

--Chris
 
All you need is a decent TC iron and some solder wick.

I prefer to tack the large ICs, particularly the TQFP fine-pitch ones down with a dot of clear nail polish between the bottom of the package and the PCB. Get them aligned perfectly and go away overnight. Perfect registration otherwise can be a real bear, particularly on the 100 and 200 lead packages. Tack all 4 corners, then add solder to all sides (a wide chisel tip works fine), then suck up the excess with solder wick. I've heard of people even doing it with a Soldapullt.

As far as the small SMD 2- and 3-lead devices, that's easy. Put a little solder on a pad, bring one lead of the component up to it, heat the solder, position the component, then remove heat. Do the other leads as normal.

Heck, I didn't even have to use my stereo microscope to do this one. If I can do it, anyone can.
 
Gents,

I now have an 8bit XT IDE card from Jeff (sent to me in working order) and installed it in my HeadStart Explorer PC.
But it doesn't boot, the headstart just boots as alway without the XI-IDE screen first.
In debug when I read out the adress, it shows the IT -IDE ROM information fine. But when trying to write it to a BIN file, that BIN file is empty.
Using DOS 3.31h.

Anyone tips and ideas?
 
Some PCs don't support DMA access to the ROM areas. Try moving the ROM data to debug's RAM area and writing it.

Have you tried running the card in another ISA-bus equipped PC?
 
gurby said:
In debug when I read out the adress, it shows the IT -IDE ROM information fine. But when trying to write it to a BIN file, that BIN file is empty.
Click to expand...
Chuck(G) said:
Some PCs don't support DMA access to the ROM areas. Try moving the ROM data to debug's RAM area and writing it.
Click to expand...
Quick example for gurby:

Code: 
C:\> DEBUG

-N MYF600.BIN          (file will be named MYF600.BIN)

-R CX
CX 0000
:2000                  (set CX to 2000, used by following write command [quantity bytes to write])

-M F600:0 2000 0100    (copy 2000H bytes from F600:0 to offset 0100 in local segment)

-W 0100                (write from offset 0100 in local segment)
Writing 2000 bytes

-Q
 
Thanks for the debug code, very helpful! I will try that soon.

I've also tried to execute the ROM directly from DEBUG, which made the PC to hang. Is it possible to move over to RAM and then try to run from RAM/local segment as a test?
 
Sorry modem7 but I don't know how to enter your syntax in DEBUG.

This is the code I used before, to try to create a .bin file:

-rcs
CS 1266
:d000
-rcs
CX 0000
:2000
-n test6.bin
-w cs:0
Writing 2000 bytes
-q


What should I try in DEBUG now ? I guess at least your F600:0 should be different value?


By the way, I don't have another PC to test the card on ...
 
Last edited:
gurby said:
Sorry modem7 but I don't know how to enter your syntax in DEBUG.
...
What should I try in DEBUG now ? I guess at least your F600:0 should be different value?
Click to expand...
What I provided was an example of what Chuck(G) was writing of - inclusion of code to copy the ROM contents into the space used by DEBUG.

I can see by the code that you have just posted, that your intention is to dump the 8 KB at D000:0
DUBUG code to do that is:

-n test6.bin
-m D000:0 2000 0100
-r cx
CX 0000
:2000
-w 0100
Writing 2000 bytes
-q
 
You must log in or register to reply here.
Back
Top