G5 Case Front Panel Audio and Fan Controller

Front panel audio sense is not wired the same on the G5 as on current MBs, if you wish to use HDA and have the front jack operational, a converter is necessary. The Mac Audio_Sense is 10k when no headphones, open when inserted, while MB FAUDIO_JS must be grounded to enable front panel jack (headphones), open for back panel jack. The left hand circuit converts the front panel signal to the MB one. I also on the same small board built a fan controller. It has a thermistor on a 10" lead to go up to the top of the case. I run the fans on 5V as 12V is too noisy and more airflow than needed for any reasonable condition. The fan goes on at about 88F and turns off a couple of degrees cooler. You can change the ON temperature by changing the 100K resistor. If it cycles too much, change the 500K feedback resistor from to 250K.

Now the front panel audio seems erratic....I will post a fix after I track down the problem....sorry I posted prematurely. Reason: the FAUDIO_JS has a big 500 Hz component that was affecting the transistor switch which was originally unbiased.

I added a filter cap on the board and a feedback cap., now it seems to work more reliably. I will test for a few more days and report. 1 uF cap. seems to work fine....no dropouts, note that the value might be different on a different MB.

Yes, the NPN transistor, resistor and cap. connect the G5 front panel pin Audio_Sense to FAUDIO_JS on the MB.

NOTE: I have updated the schematic a couple of times, so please look at the current one.

SnapMan said:

Front panel audio sense is not wired the same on the G5 as on current MBs, if you wish to use HDA and have the front jack operational, a converter is necessary. The Mac Audio_Sense is 10k when no headphones, open when inserted, while MB FAUDIO_JS is ground when headphones, open otherwise. The left hand circuit converts the front panel signal to the MB one. I also on the same small board built a fan controller. It has a thermistor on a 10" lead to go up to the top of the case. I run the fans on 5V as 12V is too noisy and more airflow than needed for any reasonable condition. The fan goes on at about 75F and turns off a couple of degrees cooler. You can change the ON temperature by changing the 100K resistor. If it cycles too much, change the 1M feedback resistor from 1M to 500K.

Now the front panel audio seems erratic....I will post a fix after I track down the problem....sorry I posted prematurely.

Click to expand...

the 1M resistor and the npn is for the audio sense?

Would you mind showing a picture of your wiring? And if I don't need implemented fan control (very nice btw), should the collector side of the npn just go to ground?

The fan controller has never missed a beat while the audio circuit has periodic dropouts. I am not sure why, but have not had time to work on it. One other poster suggested a solid state relay and I am inclined to think that would solve it as they have good isolation and low resistance in both directions.

FWIW, I attach a photo of the breadboard. You can see the .1" header at the bottom where a single connector in the case is connected to the front panel headphones switch and the mobo switch input. The 4 pin power connector brings in 5V to power the audio and fan circuits.

My intention is to get it working and then make a small board using the laser printer transfer method as I have converted a couple of G5 cases and have a few more interested in doing so.