Front panel audio; Case header has no pin 4

[EvilKitty]

My case's audio header has 8 pins populated (it is a old HP case, no older than 2003), pin 4 is not populated, since AC97 has 5 pins occupied and HD audio has 9 pin occupied, i am guessing this header uses proposed standard that was not complete, can this be made or rigged to with with a modern motherboard?

i did manage to fine this image:

should i just jump pin 4 to ground to make it work?

[thedangerine]

Jumping it is worth a shot. Not sure what the layout on that would be but likely wouldn't hurt it, just not work

[EvilKitty]

Well i just had enough fun for one night

after jumping the wire, no boot screen,

windows shows up at wrong res and cant change

clear bios

now i have nothing

swap in old GTX 600 Series gpu

now i can get into the bios!

while trying to boot several of my linux ISO files from my flash drive and nit getting to the desktop, i managed t find a error:

something about a missing EDID on HDMI one, well i know a EDID is proved from a display and to proved video config data, well if there is no EDID being givein to the PC that explains the video issue, lets try the other HDMI port, well now it works...

jump audio wire in pc  -> display will no longer provide EDID on the HDMI port i was using, what kind of black magic can use that as a cause to make that effect

 

jumping the wire to ground via USB plug had no effect what so ever

[EvilKitty]

So i have done some multimeter checking

From the boards audio jack the wires goes to a 14 pin out female plug (13 has no pin, no it is not numbered):

1  2
3  4
5  6
7  8
9  10
11 12
   14

I believe the extra pins are to accommodate a audio line in, i turned the separate lead for this into a extension to make a rear I/O passthrough as no board has such a header

My motherboard has this pin out for it's audio header:

 

From here i labeled the pin out of the case's audio cable (pin arrangement in code block above)

[pin] [name] - [MB pin]
1  L Headphone - 9
2  Detect Mic - 6
3  Detect Headphone - 10
4  R Headphone - 5
5  L Line Out
6  M Line Out
7  NC
8  R Line Out
9  R Mic - 3
10 Sense Send - 7
11 NC
12 L Mic - 1
14 Ground - 2

from there i did a connectivity check between every pin

1  to 03 when no headphone
2  to 04 when no headphone
3  to 01 when no headphone
4  to 02 when no headphone
5  to 07 when no line in
6  to 08 when no line in
7  to 05 when no line in
8  to 06 when no line in
9  to 11 when no mic
10 to 12 when no mic
11 to 09 when no mic
12 to 10 when no mic
14 Ground

I also reverse traced from the jacks to see where they go with a TRRS cable

*order is from the base of the TRRS cable to the tip (Base, Ring, Ring, Tip)

Line in (CTIA standard)
14 Ground
6  Mic
8  Right
5  Left

Mic (CTIA standard)
14 Ground
14 Mic
12 Right
9  Left

Line out (CTIA standard)
14 Ground
14 Mic
4  Right
1  left

 

I did test jumping Sense send to detect and it reads it as a jack detection

i am guessing i can just jump these 3 pins or put it on a button to turn the mic/headphones on/off as i see no way to make it work like it should

[EvilKitty]

enabling headphone disables rear audio

guess i will have to use a button toggle, unless someone knows a way to make it work, what on earth is this case trying to do with it's audio out? was it some custom audio detect AC97 system?

[EvilKitty]

So this is the wiring solution i have come up with:

Now if GIMP only had spell check...

[EvilKitty]

I ran into a issue with this design

when connecting the 1.8v send to the base of the 2n2222a it randomly enables the mic and or headphone when using 1k as well as 22k resistors

maybe i need more ohms to work around this, lets try 1 mill... well that works but the led is very dim since the gate is barley open at all, so i tried 100k it seems to work well at least with a single transistor, but that is enough to very if this will work, maybe i'll need more when using 2

[EvilKitty]

Update: Using 2 100k resistors works fine until you try to turn on both the mic and headphones and which point they both read off

hard to believe using less than taking 0.036 milliamps causes a problem

using a pair of 470k resistors does work, however the LEDs are very dim as the gate is not very open on the transistor

using 168k (100k+68k) also works, but the leds are very dim

i think i am gonna use 4 2n2222a transistors to make a pair of darlington transistors

[EvilKitty]

This is what i ended up doing, using the transistor thing was inconstant when using both mic and headphones

Button LEDs are always on while the system is running