Question about the HDaudio sense_send pin

[evilkitty]

I am in the process of rigging some old AC97 jacks or work with HDaudio (the jacks are behind a glued on face plate, so I'm not gonna swap them)
The pinout for HD audio has Sense_send and mic/headphone detect pins
so jumping sense send to a detect pin enables the jack
i also found out that pulling just a few thousandths of a miliamp from sense send causes very odd behavior like the ports enabling/disabling at random even with the detect pins being in a NC state
now i am wondering what this sense_send pin really is clearly it can't just be a 1.8v power rail as i though it was

[ratdude747]

Re: Question about the HDaudio sense_send pin

No, the way HD audio works (in terms of headphone/speaker signal routing) is completely different than AC97. Some cards support both (such as at least some of the ASUS Xonar cards), but generally they are not quite compatible with each other, and it's not a trivial swap (tried it, waste of time/effort)

AC97 routes the headphone/speaker out to the headphone jack, and then back through NC contacts in the headphone jack (one per audio channel, total of 2) to the AC97 port, which is then internally routed to the speaker jack (why such requires jumpers if not using the jack). The NC contacts are forced open by the insertion of the headphone plug into the jack (this is how most headphone ports on PC speakers work). The Microphone jack has no switches, and is a direct line to the sound card/chip; microphone presence is not detected.

The upside to AC97 is that it's a purely electromechanical solution that didn't require any changes to existing audio chipsets (merely the addition of a header to motherboards/sound cards). The downside is that when not using the headphone jack, one's audio signals pass out to the jack and back, which can lead to "excessive" noise on the output due to many AC97 cables not being shielded (or having a grounding fault causing a ground loop). Also, there is no way to automatically mute the microphone jack, which is a bit of a nuisance.

HD Audio uses the same header, only with the "speaker return" pins repurpoused as "plug detected" digital active-low (pulled up) inputs, one for the headphone port, one for the microphone port. They are driven by an NO contact in the end of each jack, referenced to the audio ground. These tell the audio chip (and by extension, the OS/sound driver) whether the jack in question is plugged in, and the OS/Driver uses such to change the audio I/O settings accordingly (usually configurable with the official driver software for the chip in question). This fixes both issues with AC97 listed above.

My understanding is that an AC97 jack set can be used on a HDA port, but with no plug detection (if anything, with nothing plugged in, it would spuriously detect jacks due to the audio signal being fed into the sense ports). The other way around is not compatible; some cases/jack sets will skirt the issue with an "AC97 adapter" plug/pigtail with the speaker feedback pins jumpered (no jack detection, both ports are in parallel and always active). Technically one could made an interface board with a dual-NC relay (and an audio-ground referenced power source), but due to no constant power available on the AC97 header (or the HDA header, for that matter) and audio grounding being often isolated from power supply ground (to prevent ground loops and otherwise reduce noise), such can't be made as a "universal" plug-and-play adapter, which is why such isn't a thing.

I did try to mod some AC97 jacks to have the needed sense switches for HDA (and modify said board to make it work), but that wasn't successful. I later was going to go so far as to make a new PCB with HDA-style jacks, but the case didn't have room for the longer jack bodies to fit. I ultimately swapped the creative X-fi card I was using for an ASUS Xonar DG card, which can be set in the driver software to emulate either standard on its header. If memory serves, the Xonar DS (one model nicer PCI card) also supported this feature. Not sure if these cards are still produced and if not, how hard they are to come by, but they are an option. There are other cards with headers (albeit "fixed" to one standard), but in my experience, they are rare (such as the aforementioned X-Fi series of cards (HDA), and some other generic AC-97 card I used to use that I somehow managed to fry) .

This answer your questions?

[evilkitty]

Re: Question about the HDaudio sense_send pin

what your post has useful information, it is not what i was looking for
I just wanted to know what the SENSE_SEND pin actually is
JAUD1: Front Audio Connector

1. MIC L
2. Ground
3. MIC R
4. NC
5. Head Phone R
6. MIC Detection
7. SENSE_SEND
8. No Pin
9. Head Phone L
10. Headphone Detection

I have found that jumping SENSE_SEND to MIC Detection or Headphone Detection causes the audio to switch to the headphone/mic out/input

by isolating these 3 wires and jumping them i can use the AC97 jacks and turn them on/off by connecting or disconnecting a jumper wire

i also found that trying to use sense_send as a power source (transistor base) this pin that reads 1.8v does not let me pull much of any power without jack detection going crazy
so i was wondering what this pin is, it must be some kind of GPIO pin, so what is it's function

[diif]

Re: Question about the HDaudio sense_send pin

SENSE_SEND is the signal coming from the HD audio IC. MIC Detection is named Sense1_RETURN and Headphone detection is SENSE2_RETURN.

https://en.wikipedia.org/wiki/Intel_...finition_Audio