Hello! I just got a new pair of (supposedly) 32-ohm headphones that sound great but don’t get very loud at all on the integrated audio board. Therefore, I’ve decided to look into building a custom DAC/amp board for my laptop. I’m creating this thread primarily to ask a few questions from people who know more about the topic than I do, those’ll be at the end.
To start off, my target is for a headphone-out jack, not a line out jack, so I can plug my headphones straight in. I’m not an audiophile per se, but I do use the audio jack for something called oscilloscope music that benefits HUGELY from higher sample rate, so 96k24 is a must and 192k would be awesome. One awesome thing about the built-in board is that it’s DC coupled, so I can have DC offset in my waveforms. This will remain the same in my board.
I am currently about 5 minutes into research into how to actually design a headphone amp that would fit in the laptop, and while I have significant experience with and tooling for doing custom circuit boards, I’ve never worked with audio circuitry before. (Though I want to learn!)
I haven’t decided whether I want to do an expansion card or a replacement for the built-in audio board, but I’d prefer to do the latter if I can figure out how to do it. So, here’s my list of questions:
Are there any physical specs for the audio board currently in the laptop?
Are there any protocol/connector specs for the current audio board? This is what decides whether I do an internal board or an expansion card.
How hard is it to design a good audio board compared to a working one, and how big is the difference?
Would anyone else be interested in this as a product if I get to that stage?
While you aren’t a repair shop, Framework would still likely provide detailed schematics to you provided you sign an NDA. That would likely answer the first 2 of your questions.
And yes, I would be interested in purchasing an upgraded board if it’s internal.
I would definetly be interested! The audio has been a pretty large deficit for me!
Isolation might be tricky to accomplish, but honestly I have no idea and have next to no knowledge the analog/audio part of PCB design either! Getting the best ICs will be very important though! I’m actually reasonably certain the DAC is on the mainboard though… I hope that the static and power state beeps are from the on-board IC…
As GhostLegion replied, Framework should be able to answer the first 2 questions if you send support an email!
Thanks! I’m definitely going to start out with that USB card, but I still want to challenge myself by cramming a powerful DAC/amp in place if the old one!
For an internal one, not sure if you can have sufficient shielding and electrical isolation unlike a desktop mainboard, in which not all implementations work well enough.
I’d say it’s worth a try. The audio board is already a separate module PCB, and an expansion card formfactor build can be shielded+isolated pretty well if need be.
I just finished a bit of research, and I’ve decided to use the ES9218PQ I²S DAC with HPA as the DAC. I still need to find a microcontroller that can do USB audio at 192k24 (or at least 96k24, and it’ll probably be a PIC32), but aside from that I have a start. I’ve also done some schematic work, basically just hooking up all the pins on the ES9218.
One small issue will be form factor, I’m going to have to fit a USB-C plug, a 1/8" jack, a PIC32, and that ES9218 onto a tiny PCB along with all the necessary capacitors and stuff. And a crystal oscillator for the audio clock.
Massive update! I just tested my audio board with an oscilloscope and I think it’s defective! Lots of noise and extremely low voltage (80mV P-P) versus my phone’s headphone jack (2.4V P-P). I’m going to replace it ASAP and see if it is fixed.
Huh, even more strangeness. When the Headphones audio device is inactive, I’m getting a 66kHz, 5mV P-P frequency on the headphone jack, and when it’s active I get a bunch of noise on top of it.
There isn’t much to break on an audio board, that’s the thing. IIRC it doesn’t have any active components, apart from the hall sensor, which isn’t involved in audio stuff.
I thought there was some kind of amplification, I guess that’s elsewhere? I’ll take apart my laptop and have a look at it in a moment. Whatever it is there is clearly something off about the audio circuitry, either by design or by defective components.
Update: partially not defective! I figured out what audio chip was driving the audio (it’s on the motherboard, a 92HD95 audio chip) and its headphone amp output is indeed specced at about that level. The noise is still an issue though, seems like there needs to be better shielding. However, I still may replace the audio board, as it’s potentially got an amplifier circuit on it from what I can tell.