IDE flash

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View

    A few years ago, I found some IDE flash ram drives that were the size of
an 40pin header connector (IDE).  To install then you just pushed them in
the ide slot directly, and it was ready to go.

    Anyone know of a source of these today?  I was never able to purchase
them, and the 2.5 & 3.5 FF drives are to big.  I don't need one bigger than
8 Megs.


Re: IDE flash
Quoted text here. Click to load it

Not exactly the same thing but you can get adaptors to use a Compact Flash card
in IDE mode.

Re: IDE flash
On Fri, 11 Jul 2003 18:19:21 GMT, "C Wood"

Quoted text here. Click to load it

I think your best bet might be to get IDE<->CompactFlash adapters.  
Alex Pavloff - remove BLAH to email
Software Engineer, Eason Technology

Re: IDE flash

Quoted text here. Click to load it

The only one I know is Disk-on-Chip IDE Pro:

Roberto Waltman

Re: IDE flash
On Fri, 11 Jul 2003 18:19:21 GMT, "C Wood"

Quoted text here. Click to load it

Disk on Module from the same folks that bring you Disk On Chip. A
flash based disk emulation with a built in IDE connector. I have seen
several recently, and believe they still sell them.

Bob McConnell

Re: IDE flash says...
Quoted text here. Click to load it

I really don't know why ANYONE would want that Disk-on-a-Chip/Disk-on-a-
piece-of-corn-on-the-cob stuff.

CompactFlash in TrueIDE mode works like a charm, it comes in many sizes,
it's "second sourced" given there are many companies making CF cards,
the specs are available for free, it's easily transportable, readable by
laptops, desktops, etc, etc.

I had a friend at a company who was trying to evaluate the Disk-on-Chip
stuff and the company basically ignored him since his quantities were
less than 1K/year.

Anyway, I think as others have suggested, CF is the way to go!

-- Jay.

Re: IDE flash
Quoted text here. Click to load it

Sounds like you've used a CF in true IDE mode??  I am in the midst of
 so in an embedded project , but am running into walls.  Do you by
chance have an exmaple code I can look at?  Thanks.


Re: IDE flash
Hi Tim, says...

Quoted text here. Click to load it

I have to dig around and seee if I have any schematics left from the

If you get the CF spec( (which I am sure you have)
it tells you what happens to each pin in each of the modes. Typically,
for TrueIDE, if memory serves, you have to more or less pull down pin #
39. You do have to follow the spec and pull-up and/or pull-down some of
the other lines too, but they're not the "critical" ones.

Send me an e-mail with your address and I'll see what I can do.

-- Jay.

Re: IDE flash

Quoted text here. Click to load it

Any chance you can post it to the group?  I'm working on
a similar problem.

Re: IDE flash
Quoted text here. Click to load it

Will do! Many people here have helped me out, this is the least I could

Note 1:

I made this table when I was trying to get the CF card to work in
TrueIDE over the PCMCIA port.  All the information still applies, but
note that the "functionality" that pin 39 provides on the CF card isn't
found on the PCMCIA port (i.e. CF pin 39 doesn't mate to any pin on the
PCMCIA port even with a CF to PCMCIA adaptor).

To solve my problem (after making this table), I got a CF to PCMCIA
converter and opened it up and hacked it, so I could get direct access
to signals(like pin 39) and lift pins when needed. This hacked adaptor
DID work reliably in my application, the host treated the CF card as a
TrueIDE device (over the PCMCIA bus!).

Note 2: Signal direction is from the perspective of the CF card

Note 3: Many of the pins listed in this table were suggested to be
disocnnected from the host by the CF spec(when the card is in TrueIDE
mode). The host MAY pull-up or pull-down some lines, which might mess up
the card, even if you have the right polarity. Be careful of this,
that's why I suggest you disconnect them all if you can.

Note 4: CF spec says pull-ups should be >10% kohms and pull-downs should
be >= 100k ohms. I suspect this is for power reasons, but be aware...

Here we go:

CFPIN    PCMCIA_PIN    Direction    Pull-Up/Down        Name
8    8        I        Pull-down        A10
9    9        I        Pull-down        -ATASEL
10    11        I        Pull-down        A09
11    12        I        Pull-down        A08
12    22        I        Pull-down        A07
14    23        I        Pull-down        A06
15    24        I        Pull-down        A05
16    25        I        Pull-down        A04
17    26        I        Pull-down        A03
36    15        I        Pull-up        -WE
39    (None)        None        Pull-down        -CSEL
43    60        O        (Nothing)        -INPACK
44    61        I        Pull-up        -REG

Important signal description:

Pin 9(-ATASEL):

This is really "the" signal that tells our CF card to goto TrueIDE mode.

Pin 39(-CSEL):

This tells the CF card, in TrueIDE if it's Master or Slave on the IDE
channel. This signal is internally pulled high(Slave), but to make the
card the Master on the IDE bus, pin 39 needs to be pulled-low(Master).

Since I did not have the schematics to the PCMCIA host I was working
with, I did not know if they pulled certain signals high or low (i.e.
violated the PCMCIA spec). For this reason, I suggest, at least for
debugging, you disconnect all the above pins (from the host to the CF
card) and make sure the above CF pins are tied high or low through a
resistor to be safe.

Also, if this applies, make sure that your PCMCIA Card Detect pins
(1,34,35,58) are grounded. When messing with my adaptor, I missed one
and my host wouldn't recognize my card until I fixed that.

I have a PDF, which I created from reading books, specs, that actually
makes a mapping between IDE and PCMCIA with pin names, etc. I'll
consider sending it out. (I have a PDF of the above table too, but I
don't think people here take kindly to binaries). For what it's worth,
I've also hooked an IDE HDD to the same PCMCIA host that used my CF card
in TrueIDE mode!

Let me know how it goes!

-- Jay.

Site Timeline