HARD DRIVE ACTIVITY LED

[Bobby]

Hello, All!

Could someone tell me which pins on the IDE interface provide hard
drive activity? Also, what specific LED would you recommend?

Thanks!

Bobby

 

[Don Taylor]

Hello, All!

Could someone tell me which pins on the IDE interface provide hard
drive activity? Also, what specific LED would you recommend?

Thanks!
Bobby

It is a little old but you might still be able to find an copy
of "The SCSI Bus and IDE Interface" by Schmidt, Addison Wesley.
Unfortunately, probably due to legal constraints, that book doesn't
give complete information on how these work. The official version
of the standard, which reading can be "like swimming through tar on
a spring afternoon" you have to pay to see, but you can get some some
additional information and maybe figure out answers via

http://www.t13.org/

But the book gives you a good overview of the way that the IDE
interface works and helps you get started with this subject.

There is no single pin that turns the LED on and off. Instead, the
interface passes commands (with or without accompanying addresses
and/or blocks of data) back and forth between the computer and the
drive. These commands are things like "read a block of data of
this many sectors starting at this address." And if the drive is
busy carrying out one of these commands then the drive turns the
LED on.

I've thought now and then about building a little card that would
plug into an unused connector on the IDE cable. This card would
decode the commands and addresses and use this information to drive
leds and give you a visual indication of what was going on with the
drive, but I haven't had the energy to get that built.

 

[Paul]

Hello, All!

Could someone tell me which pins on the IDE interface provide hard
drive activity? Also, what specific LED would you recommend?

Thanks!

Bobby

There is some info here.

http://www.pjrc.com/tech/8051/ide/wesley.html

What I cannot tell you, is if there is any unintended
consequences of picking off the signal as the
author of that web site has done. Caveat emptor.

The author of the above web page, used a 330 ohm resistor
and a LED. The current flow would be (5V - 2.0V) / 330 =
about 9mA of current. The 2.0V is a rough value for a red LED,
and other color LEDs use a bit more more voltage than that.
9mA should be plenty for a high efficiency LED. Remember
that a logic chip on the disk drive has to sink that 9mA
current, so don't force it to sink more than that. In
other words, try a 330 ohm resistor or one with higher
resistance, like 470 ohm, 680 ohm and so on.

Radio Shack sells LEDs and resistors. That should be a
good enough source for the job. There are specialized LEDs
available from obscure sources, but that would be more
trouble than it is worth. If you change the color of
the LED, recompute the above equation, taking the new
value for the forward voltage drop into consideration.
(That is the 2.0V in the example above.) That is where
a LED product that you can get a datasheet for, comes in
handy.

Note that the forward voltage drop for the LED, is actually
a function of the current. My picking the value of 2V
is just an "engineering guess" intended to arrive at a
rough value for the resistor quickly. You may refine
your calculation by using the curve in the datasheet, if
you want. (Fig.2 here shows about 1.9V at 9mA)

http://www.liteon.com.tw/OPTO/SPEC/DATABOOK.NSF/PASN/DS20-2002-243/$file/LTL81HKEKNN.pdf

DASP- is pin 39 in the T13 document (while the above web
page referred to it as /ACT). 12mA drive is available
on DASP, of which some current would be used by the motherboard
pullup, if provided.

http://www.t13.org/docs2002/d1532v2r1a-ATA-ATAPI-7.pdf

Try your first experiments, with some old drive you don't
use anymore. Just in case

Also, realize that overloading the drive capability of
DASP- may take time to show an effect. I remember a guy
I worked with many years ago, overloaded a driver like the
one inside the disk drive, with 3X the normal current. It
only took about two days of running, before the chip was
killed And that is why we don't want to make the LED
too bright - potential overload.

Paul