EE resources for audio hardware design?

[The Bill]

My Google-fu has been failing me on this. I'm looking for a good variety of online resources and textbooks on the design of audio equipment.

The level I'm looking for would be for upper-level undergraduate or intro graduate level electrical engineering. Topics must include how to best isolate inputs, outputs, and power inputs from line noise, ground loops, RF noise, crosstalk, preventing unintentional creation of RF noise, etc. In-depth coverage of the practicalities of designing systems with ADCs and DACs is also necessary.

The searches I've tried so far mostly give results related to audio engineering, and hardly anything to do with designing hardware. I've got quite a few general EE texts, but none that focus specifically on audio frequency equipment and catering to the needs of audio professionals and audiophiles.

 

[BiGyElLoWhAt]

Pretty much all audio equipment comes down to filters and modulators.

I'm not sure what you mean by isolating inputs, outputs, etc. I think a lot of the noise would be due to the fact that wires act like antennas.

There are different types of noise:
60hz (comes from the fact that our wall power in USA is 60hz)
Spectral (i.e. frequency dependent, white noise, brown noise, etc.)
Shot (similar to single electron noise, should be irrelevant)
and others

Probably what would be useful would be looking into different kinds of noise and seeing which ones would be relevant for your design.

You'll want to be familiar with IC's (op amps, etc.) and their uses, particularly as amplification.
LC - notch filters, band pass filters
RC - high pass, low pass filters

These are the big ones that I can think of. I know it's not a book, but these are some of the more (obviously, to me) useful circuits and design elements.

Diodes and potentiometers (pots) are also commonly used. Pots are largely used in conjunction with op amps for variable gain.

 

[DaveE]

One of the best EE textbooks for this sort of thing:
"Design With Operational Amplifiers And Analog Integrated Circuits" - Sergio Franco

 

[DaveE]

Also this one is worthwhile:
"Troubleshooting Analog Circuits (EDN Series for Design Engineers)" - Robert A. Pease

This one is free and good, but the ICs they refer too are all old, there are better ones nowadays:
http://bitsavers.trailing-edge.com/...aBooks/1980_National_Audio_Radio_Handbook.pdf

 

[berkeman]

Topics must include how to best isolate inputs, outputs, and power inputs from line noise, ground loops, RF noise, crosstalk, preventing unintentional creation of RF noise, etc. In-depth coverage of the practicalities of designing systems with ADCs and DACs is also necessary.

I've written some chapters in our company's User's Guide documents to help our customers design our network transceivers into their products and avoid EMI issues (and desensing of RF transceivers due to noise coupling). I'll look at them again to see if it would help to point you to them.

One of the biggest keys is to think in terms of "Star Grounding" where you do your best not to share ground (or power supply) impedance between different circuits. The last thing you want is to have digital microcontroller and memory noise coupling into your sensitive analog circuits. You also use the Star Ground concept to help lay out your ESD/Surge protection for inputs, so that any transients conducted into your device are shunted right away out the power supply or other grounding areas.

Have you seen these concepts before, or is this part of what you are wanting to learn? I've seen some of this in textbooks for high-speed digital design and for RF device design, but I'll have to do some digging.

 

[berkeman]

Here are a couple classic texts that are somewhat related to your questions. They are not cheap, but should be available at your local technical library (worst case via inter-library loan) for you to look through:

A great text written by one of the first founding engineers of a company that I worked at for many years:

https://www.amazon.com/dp/0133957241/?tag=pfamazon01-20

A classic and fundamental text on protecting your device from external transients and other issues:

https://www.amazon.com/dp/0486425525/?tag=pfamazon01-20

 

[DaveE]

I 100% agree with @berkeman's comment about grounding. This is really important.

At audio frequencies, with good design, you can also make high CMRR differential amps. If you think of every signal as differential, even if one side is "ground" it will inforce the concept that "ground" doesn't really exist in very low noise or high precision circuits. In any case good analog systems have multiple grounds defined in my experience. At a minimum I would start with Power Ground, Digital Ground, and Analog Ground; i.e. separate everything unrelated from each other. In particular, Analog or Signal ground should not be expected to carry significant currents, it is JUST a voltage reference.

 

[TonyStewart]

In order to design anything , you need to appreciate , product specs, and design specs are different which include all the dependencies with measurable parameters to achieve all the customer specs which includes regulatory EMC, and safety. This takes reading a hundred different product specs , maintenance manuals and DVT reports if you find any (which verify design specs with tests) and comparing price and performance.

EMC is a different kettle of fish. The best book was Henry Ott's for me. ( < 500 pages) link below

After studying a logic and analog design then study physical design, from every component choice there are many reasons. This may require studying a hundred teardowns of commercial HP & Tek test equipment manuals and equipment to Yamaha receivers. Then try to figure out the reasons for every component choice and enjoy the beauty of simplicity and complex solutions to achieve excellent performance. There is a broad range in cost/performance and design takes many years to master after you learn from the best.

Filter, control theory, and Impedance control with gain and CMRR skills are valuable to learn because you will have noise and instability otherwise.

I do not have a great library to give you but here are some examples

https://www.ti.com/lit/eb/slyt701/slyt701.pdf
https://www.onsemi.com/support/design-resources/appnotes
https://www.onsemi.com/support#downloadable
Everything you wanted to know about electronics in small modules.
http://electrical-engineering-portal.com/download-center/books-and-guides/electrical-engineering-3
https://www.yuden.co.jp/resource/data/basics/basics_en.pdf < Low ESR Caps , High SRF L

Most of all you must be creative. Like Steinberg, who use Yamaha Audio designs. Don't be afraid of the daily choices to make/buy and ask for advice.. It is essential not to reinvent the wheel unless you can profit by it. ... and you aren't asking to design the transistor but you must learn how to read datasheets to know how to find the best bang for the buck that is available.

You ought to learn a few tricks from Who the heck was Bob Pease?
http://www.apple1.fr/student_manual_for_the_art_of_electronics.pdf

A few more archived free books
https://archive.org/details/highspeeddigital0000john?q=henry+ott+emc
https://archive.org/details/handbookforsound0000unse_d9v1?q=henry+ott+emc
https://archive.org/details/fe_Op_Amp_Applications_Handbook_Newnes_Walt_Jung
https://archive.org/details/electronicsystem0000barn
https://archive.org/download/electroarch2020 (lots of stuff)
https://archive.org/details/manualzilla-id-5653942/mode/2up power design
https://archive.org/details/practicalanalogd0000unse/page/n381/mode/2up < complete design borrow 1hr
well you get the idea. Its not all in one place. But that's my fun-goo search in 10 minutes.

 

[The Bill]

What I'm looking for is references that could be handed to an electrical engineering EIT who hadn't worked professionally with audio equipment to help get them up to speed. Ideally, after working through the material, they'd feel confident at an entry level in any part of a project like, say, designing a USB audio interface with inputs and outputs.

Preferably focusing on the audio/analog side of the circuits, since they'd probably be pretty familiar with digital circuit design if they'd graduated anytime in the past 30 years or so.