From: email@example.com (Eric)
Subject: Re: PCMCIA -> CF -> ATA Adapter (A question on addressing in CompactFlash and PCMCIA)
Date: 1 Jan 2003 04:42:40 GMT
Organization: Your Organization
Abuse-Reports-To: abuse at airmail.net to report improper postings
NNTP-Posting-Time: Tue Dec 31 22:42:41 2002
NNTP-Posting-Host: !b9V'1k-WR[HZuj (Encoded at Airnews!)
X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit)
>There are adaptors around that do this, letting you plug an IDE drive
>into a PCMCIA card-slot.
>(Google search: "PCMCIA to IDE", including the double-quotes.
>However, I haven't ever seen one that was under your price point of
>US$20... most of them have been more around US$50 - $100.
>That said, I think that the WWW page which you noted in your posting,
>above, makes things pretty clear. It appears that there are a number
>of pins which PCMCIA supports which IDE does not - the address pins
>you mentioned. I'd imagine that, since they are output pins, you
>don't need to do anything with them. There's a number of IDE pins
>which are marked as "GND" which should be grounded, and then there are
>the IDE signals DMAREQ, SPSYNC, DMACK, which I take to be "DMA
>request", "Spindle Sync", and "DMA Acknowlege", respectively. It
>appears that the PCMCIA interface doesn't really have a way to use
>these pins. It is possible that one might contrive to hook them to
>the unused PCMCIA address lines, but that would appear to me to be
>Doing a tally of the IDE lines listed in that WWW page, all 40 are
>accounted for, so mechanically, it looks like a pretty straightfoward
>connection - just wire them straight through, as the signals all seem
>to be the same type... PCMCIA and CF were designed with the IDE
>interface in mind as a model, after all.
>Software is going to be the tricky part. You may need to write a
>driver for this - I have my doubts that the OS inside your laptop will
>be able to talk to the IDE drive without a driver. Perhaps a starting
>point would be the PCMCIA-cs drivers for Linux. See:
>Really, though, it looks like it's not going to be too terribly hard
>to do this project, judging from the looks of the commercially
>available equipment to do the job.
>I'd have to say that thinking about CF here is getting you off track:
>what you're really doing is connecting an IDE drive to a PCMCIA port.
>If you happen to use a PCMCIA/CF adaptor as a hardware shell to make
>building the connectors easier, well, so be it... you'll only need to
>be concerned with finding which pins of the PCMCIA side come out where
>on the CF side... so that you can wire them to the IDE side. I'm
>thinking that perhaps you should disassemble the PCMCIA/CF adaptor,
>and remove the CF socket - that should allow you to solder the
>conductors of a standard IDE ribbon cable in its place, which could
>then be run to the drive.
>Hope this helps,
>Anthony J. Albert
Well you'd be correct if the PCMCIA slot only listened to a specific address,
but I'm not sure if that's the case or not. If it is, then I'm set. Maybe that
would be a better more revised question to ask. However, if the PCMCIA slot
listened to more than one address, then I'd have to put the MSB of the Address
lines through a decoder and AND it with the /OE signal right?
I think you're right about it being straight-forward and that is what prompted
my inquiry. Thank you for composing a well thought out response. My alma mater
carries a book on PCMCIA and I'll see if I can browse through it and get the
answer I'm looking for.
What I really need to do is put a logic analyzer on the thing and see the