hdd access leds

[phreak97]

does anyone know exactly how to add access leds to hard drives other than the primary? i located the access led pin on the cable, but what do i do with it? should i just run it through an led to ground? or just cut the line and run it through an led then let it continue to wherever it goes?

any info?

[Aidan]

Normally, the motherboard mixes the signals from the IDE headers together to provide a single HDD access LED. As far as I'm aware, the LED output signal on the IDE cable isn't used for anything else.

You can cut the signal out, wire it to a suitable resistor, then onto the LED, you should be ok. Remember that the signal on the IDE cable represents the access signal from all devices on that cable (if you happen to have two hard disks).

[phreak97]

thanks, ill try that out tonight. yeah, i know it'd be both, so ill cut it after each device, that should work ok shouldnt it?

[ilesj]

Some (or most, don't really know..) hard drives have pins in them for the access LED. Check your drives if they do.

[phreak97]

they dont, only via the ide slot

[Aidan]

How did you get on with this?

[phreak97]

with no resistor at all, the signal seems a little low, the led doesnt light terribly well, and it doesnt seem to flash at all the right times.. i got some free quad comparators, so i might se if i can get one of those working on it. probably overkill where i could just use a transistor, but the comparator is actually more compact than a transistor anyway.. soldering will be fun :s i might make up a little pcb to accept 4 channels, they can all run off one comparator (being a quad (4 channel) comparator.

comparators are the chips which control those little led bar graphs on pre digital display stereos, you give it a voltage, and if the input signal rises over that voltage itll turn on the output signal (fixed voltage i think), then if it goes below, itll turn it off again. there are 4 of these in a quad comparator, so you can input 4 target voltages which will set the level at which the outputs activate, so it can run 4 leds.

ill just put a pot on the input, then adjust it untill i get the level i want. this will cover the lack of voltage, and the wrong flashing times (if they were caused by the low voltage), i hope.

[Aidan]

If the current really is low, but the voltage fine then you can get away with just about any logic gate.

When you say it flashes at the wrong times, what do you mean?

[phreak97]

it seems like it's sometimes on when it isnt reading the drive, and sometimes not on when it is reading. but this could be because the drive is taking more current when it's doing an intensive read, and when it's doing light reads the led comes on? sometimes itll work, sometimes not, but always very dim. ill get around to the comparators sometime today.

[Aidan]

/me whistles, whilst looking the other way.

It's an active low signal. That means the drive pulls it low when it's performing an operation. It's also probably an open collector output, which means that the drive is expecting to see a pull up resistor.

You could try this before you build anything: Calculate a sensible sized resistor for your LED. Connect the resistor to +5V, and the LED. Connect the other side of the LED to the ribbon cable. Try to keep the current through the LED low, as I'm not sure what the specs on the DASP signal are!

[phreak97]

sounds scary, im not going to fry out a western digital 80gb am i? (ill probably try it anyway)

[Aidan]

Basically, the output pin on the drive works the following way:

When the drive is not active, the pin is 'floating' (actually, there's probably a weak pullup to bring it near to 5v). When the drive is active, the drive grounds that pin.

Hence, if you connect the LED/resistor combination to the +5V line, the drive can pull the other side of it to ground when it's active. This way both drives can work the same line without fear of shorting each other out.

If you keep the current low (ie, just enough to light the LED), then you should be ok.

[phreak97]

well.. the leds like 20mA is that going to be too much?

[viletim!]

this should do the trick

the value of Rs is just whatever you'd normaly use in series with the LED, 390 ohms for a standard red led.

Code Select Expand

                 * +5v
                 |
                /e
              |/
IN o---/\/\/---|    BC558
       10k    |\
                \c
   LED          |
+---|<--/\/\/---+
|        Rs
|
--- GND

[phreak97]

ill try that out in a min thanks

[phreak97]

it's times like these i wish there was a 24hr electronics store.. i only have a BC548 and it's 7pm. ill look for a used one:P

well.. it's doing strange things, but ill try again in a more reliable pc later

[Aidan]

*DASP has to be able to sink at least 12mA, and it's pulled up at the device end with a 10KOhm resistor (according to the T13/E00120R1 spec).

Viletim's circuit will not work in the way you want it to, as it's expecting the line to be active high (IE, drive activity is indicated with +5v). Instead, the drive grounds the line when it's active.

Grab yourself some NAND, NOT or NOR gates (as I presume you have more than one drive), and use them. (or, use two transistors, and have the first cut off the drive current of the second.)

If you use NAND or NOR gates, just wire the two inputs to the gate together, connect your LED/resistor to the output, and the *DASP line to the input.

[viletim]

Aidan,
I think my circuit will work ok. I should have mentioned that the BC558 is a PNP transistor. They pretty much grow on trees here but from what I've heard they're pretty rare in some places (like the states).

When the input is at +5v or Hi-Z, no current flows from the base and the transistor is off. When the input is grounded then current flows out from the base and the transistor is saturated, letting current through to the LED.

It's not an ideal circuit though. If the input is at, say, 3v - a perfectly legal TTL high, then the transistor will still saturate and turn on the LED, maybe a higher value base resister will help in that case. Of corse, as you say, logic a gate is required for any guarantee.

But not all logic gates are very well suited to driving LEDs. A 74LS can drive a standard led as it can sink 8-16mA (depends on the gate). They can't source much though - only about 1 mA. If you need more current than that (eg. for a high brightness LED) then a 74AC/ACT or a transistor driver circuit is required.

phreak97,
a 2n3906 or BC559/6/7 will work in place of the BC558

[phreak97]

i found the transistor, but it didnt have the effect i was looking for, i want it basically, when connected to the primary drive, to mimic the regular hdd led perfectly.
i tried putting 5v through an led then through an 82ohm resistor, then to the access pin. which actually seems to have worked quite well. but i would like some confirmation that  it wont damage my drives having ~20mA going into them. it should be ok though shouldnt it?  

[viletim!]

hmm...ok
Try a comparator, a good choice is the LM311 - it can sink up to 50mA and it's pretty cheap.

It's not a good idea make the hdd's output sink 20mA. It wasn't designed for that much current and it might even damage it.

Code Select Expand

.                                  +5v *
.       1k       1k                    |
.   +--/\/\/-+-/\/\/-+-----------------+
.   |        |       |                 |
.  ---GND    |       |Vcc              |
.            +-----|-\     LED   Rs    |
.                  |  >----|<---/\/\/--+
.     IN o---------|+/
.                    |
.                   ---GND+Vee

[phreak97]

yeah i have some lm339 quad comparators, but so far i've been too lazy to get around to that.

edit:
omg thankyou!! the comparators work perfectly. and being a quad comparator, i can run all 4 channels off one chip:D