help-with-audio

[echo hedge]

ladyada did the reverse engineering on it

mainly through decompilation I think

[versed torrent]

that's what I was going to start with if I could find the GCC version... but seems the links to it are all stale based on the old Amtel site and the Microchip site is an impenetrable fortress

[echo hedge]

try this https://start.atmel.com/#examples/SAMD21XplainedPro/touch

the compiled lib is in the example zip

[versed torrent]

Aha - okay - might take a look at that in a bit.... At the moment I've got the other version up and running and back to making my "Pandora's Drumming Box" instrument.

[bronze locust]

I'm trying to help my parents choose a microphone for taking Zoom meetings where they are both calling from a single laptop.

• Any recs?
• What should I be searching for? My current search term is "best budget studio mic"
• Am I correct that pretty much any mic will be better than the laptop's built-in one, even ignoring the actual mic specs, since the built-in mic is right next to the speaker so there is heavy reliance on software-based filtering to remove the feedback?

[random bone]

The Blue Yeti and its siblings are well regarded. But I have a Logitech C920S webcam, and its builtin microphone is fine. It is also a better camera than is present in many laptops.

[random bone]

^

[bronze locust]

thanks @random bone! I have been using the C920 and C922 for years, but always assumed the built-in mic was junk

[random bone]

it is not as good as something like a Blue Yeti, but it seems OK. If you listen to the weekly CircuitPython meetings (available on YouTube and elsewhere), my mic is the Logitech; Kattni and Scott have Blue Yetis or a Yeti Nano

I have a Zoom H2n standalone recorder, which I can use as a good mic, and it does sound better, but it is more work to set up.

[bronze locust]

I suspect the geometry of these two-person calls will make the requirements pretty different. They sit on a couch with the laptop on the coffee table (mic can be put on any available surface, but I think coffee table is kind of it). So it's two people, each at a few feet of distance, not exactly head-on but off a little to the side.

Being an absolute noob, seems like either a cardioid or omnidirectional mic could work for this?

[glacial spruce]

Omnis or twin lavalier mics are easy to set up. The advantage to cardioid pattern mics is they have a deep null you can aim at sources of noise (like computer fans)

[bronze locust]

ah, as in two lapel mics that get mixed in hardware and present as a single 3.5mm input to the laptop?

[odd oasis]

Snowball is another option if you need a cardioid mic. Cheaper than a blue yeti, and audio quality is pretty solid.

[glacial spruce]

There are dual mic sets available that basically work like that.

[bronze locust]

it seems like everyone likes the Blue Yeti, and below the $80 price point, there's not that much agreement on which mics are good

[versed torrent]

Interesting audio observation on a Circuit Playground Express: I'm generating audio via the on-board DAC. When powered by my computer, there is a fair bit high frequency noise on the GND (thanks to computer's USB power...) and now that I have a small opamp between the board and my audio system it is fairly evident......

The amount of noise is proportional to the brightness of the NeoPixels I have lit!

When powered by battery, no problems....

[golden lark]

I have noticed that even when not playing audio. When I set all 10 of the neopixels to 200/255 brightness in Makecode without playing any audio I hear a noise like that. When I connect Adafruit's STEMMA speaker the buzzing is louder. (I discovered this after I hooked up the speaker and plugged the board into my computer to change some code. The old code had all 10 leds at 200/255 brightness.)

[median hemlock]

You might try this, it has worked for me in a bunch of places: https://github.com/adrianfreed/FastTouch

[versed torrent]

Similar to the approach that CircuitPython uses... but CP uses an external pull down resistor (1MΩ). I duplicated the CP method - and been working a charm - and in 20 lines of code!

(amusingly, I know Adrian Freed from a very long time ago!)

[bronze locust]

I am trying to piece together what happened in an A/V adventure I was not present for. I think at one point they connected the 3.5mm headset jacks of two devices together with a 3.5mm TRS stereo cable. What should happen in that case, if it's even specified?

[last prawn]

Did something break? It really depends on the device and what the headphone jacks are usually doing (aka - did they have power running through them)

[bronze locust]

I don't think anything broke. One device had audio playing in the output channel, the other I think didn't.

[versed torrent]

I'm not sure what they did - "two devices" - what two? Two headphones? Two phones? In general, connecting two outputs together is "a bad idea™". On the other hand, trying to drive two sets of headphones from one output will work, but probably be quiet and might sound bad.

Oh - two devices with "headset jacks" - Okay, so two outputs.

Yeah, not a good plan! Depending on the device's circuitry, you could end up damaging one or the other device, or both could be okay... no single answer really depends on the devices in question. BUT, the one thing that you can expect: Nothing good can come of this arrangement! (There is such a thing as passive mixing - but you generally can only get away with this with outputs designed not to drive a speaker or headphone, and hopefully with outputs you know are designed to handle to it, like modular synth outputs.)

[bronze locust]

this was definitely not an intentional attempt at passive mixing 😆

thanks for the info @versed torrent!

[grand lance]

https://learn.adafruit.com/assets/3730 what are the g pins and how do i use them

[glacial spruce]

Those give access to the "gain" input on the amplifier chip directly, and via a 100k resistor. Normally you don't need them and can use the gain jumper block to set the gain.

They can come in handy if you want to control the gain with software by hooking them to GPIO pins

[grand lance]

ok. and gain controls what? im unfamiliar with it outside cb

[glacial spruce]

It's a way of expressing how much an amplifier amplifies. If you have a low level input signal and want it louder, you'd set a higher gain, for example.

[grand lance]

so on this 18 is high and 9 is low?

[glacial spruce]

Yes, you can set it as low as 6dB or up to 18dB

[odd oasis]

If you go to the specifications section of the learn guide overview, there is a link to the schematic that details what resistor configurations result in what gains.

[glacial spruce]

In case you're curious

[grand lance]

and thats on the pins?

[glacial spruce]

That's the G and G* pins

[grand lance]

would a dual linear pot work on it?

[glacial spruce]

There's also a jumper block for if you want to just set a fixed amount of gain. They both do the same thing, just give you access to do it whichever way is convenient for you.

This sort of versatility and convenience is one of the things I like about AdaFruit designs, but I can understand how it can be a bit confusing.

[grand lance]

let me correct this because i didnt read the schem fully. would a linear 100k pot work in place of the resistor

[glacial spruce]

No, that's not how that chip works. You could wire up a 100k pot as a variable attenuator (like in the diagram you linked to), but the chip itself simply offers 5 fixed amounts of gain, selectable by hooking up a pin in various ways.

[grand lance]

so i would need a rotary selector switch not a pot

[glacial spruce]

Yes, or you could use a few individual switches, or a couple of GPIO pins if you have a microcontroller available.

[grand lance]

i do, ill have to see if i have enough pins. i do have a gpio expander so i may be ok

[glacial spruce]

All that is only if you want the ability to change the gain on the fly. Otherwise, just set the jumper for the gain you want and you're done.

[grand lance]

i only have 6 gpio pins left on my feather. i know i need at least one of those for what im putting in currently

depending on what else i put in it will help me decide how to handle gain

[grand lance]

i kinda dont get this chart and how it is applied

[versed torrent]

@grand lance - it means that there is this one pin, Gain:
If you leave it disconnected ("Floating") you get 6db of gain.
If you connect it to +V ("High") through a 100kΩ resistor, then you get 9db of gain.
If you connect it to +V directly, you get 12db of gain.
If you connect it to GND ("Low") through 100kΩ resistor, you get 15db of gain.
If you connect it to GND directly, then you get 18db of gain.
Yes - the table is not really as clear as it could be.

This scheme is a way for the chip to let you pick one of five available gain settings with only one pin. The board makes this easy by having a set of jumpers that can connect that GAIN pin in any of those five ways. Note that the chip does not support arbitrary gain settings, just these specific 5.

The intention of this gain control is not to be a volume setting controllable by the application or user - but to just the range of the amplifier for the nominal level of the input. If what you need is a volume control, you'd achieve that in the system that is sending its output to this amplifier board.

If you really needed to be able to control this gain setting programmatically, you could do by connecting it to two pins on a MCU, one directly, the other through a 100kΩ resistor. The board has two pins G and G' - the first is the Gain pin, and the second is Gain through a 100kΩ resistor... so you could just connect these two pins to the two MCU GPIO pins. — Then, by setting both GPIO pins to inputs, and setting whether they are floating or pulled up, or pulled down - the MCU can control the Gain. You'd need to be very careful to set both pins floating before setting one or the other pulled up/down. I wouldn't design a project this way if I could avoid it.

(sorry for all the edits -- I'm done now... I think it is now clear)

[grand lance]

ok. so the 100k resistor is built in and pins out to the second gain pin while the other is raw

[versed torrent]

right.

and you'd only use those two pin (G and G') if you weren't using the on board jumpers to pick a gain.

[grand lance]

ok. i may add it in and put it under misc settings after i have all the other things pinned out. for sake of testing and sett up ease, ill just use the jumper until then

right now im juggling interface image, buttons and mp3 playback concurently as i need to be able to see, select, and play the test track. unfortunately this is a head ache

[versed torrent]

I don't know what you're feeding this amp with - but 3db steps are rather coarse - and in general, you'd have some way of controlling the volume of the signal going to this board. And - so this setting is to just set the amplifier to handle the general level of signal you're feeding it. Jumper is probably the way to go if this system is going to just be playing back MP3s.

Are you decoding the MP3 on the feature MCU and playing via the MCU's DAC?

[grand lance]

feather to this: https://www.adafruit.com/product/3678 to the amp

[last prawn]

mzero's comments sound right - Adjusting the gain digitally before outputting to the i2s decoder is probably how you want to control the volume - thought you could do something in between the 2 as well

[worthy glacier]

quick question (please ping me if you are gonna respond) if i have a microphone module and if i tie the digital output with a audio jack (and ground) will it work? or is it a lost cause?

[solemn flint]

Depends on the type of microphone. With an I2S digital output, that definitely won't work. With a PDM signal, you'd want a simple low-pass filter as well, but that would mostly work to get you an analog audio signal from it.

[worthy glacier]

ok cool to know thanks!!

[grand lance]

https://learn.adafruit.com/adafruit-i2s-stereo-decoder-uda1334a/circuitpython-wiring-test for this board: the row of pins on top, how do i use them and can they be wired to gpio

im particularly interested in mute, though i have an alternative in mind if i cant

[solemn flint]

The Learn guide does explain what they're used for: https://learn.adafruit.com/adafruit-i2s-stereo-decoder-uda1334a/pinouts

[grand lance]

yes. very briefly. i dont understand them entirely though and it includes nothing for actual use

[solemn flint]

For more details you'd probably want to refer to the datasheet of the chip it uses.

But the MUTE pin specifically sounds pretty easy: set to logic high to mute the audio.

[grand lance]

can i do that through a gpio?

[solemn flint]

Yep.

[grand lance]

then id have to link it to a button, which im having trouble programing

[solemn flint]

You could also wire the button directly to the pin if you wanted, with the other side connected to 3.3V.

[grand lance]

sorry. virtual buttons

though that is my backup

[grand lance]

so to make sure i understand: high is power pins, low is ground

[solemn flint]

Well, high is anything that reads as a logic "1" to a digital input, which you can get either from the power supply or from a GPIO set to output a digital "1", etc. Ditto for low: could be a direct ground connection or a GPIO set to output a "0", etc.

[versed torrent]

Note - you would almost certainly NEVER programmatically control the other signals there from GPIO - because they don't make any sense to be user or runtime configurable. SF1 and SF0 set the format of the serial data. If you are using I2S - just leave them disconnected. Unless you are integrating with video equipment you will have no use for SCLK and PLL. Lastly - De-Emphasis is something you'd only want if you were specifically playing audio that has been pre-encoded with Emphasis. If you are playing audio files this is almost certainly never the case.

Also - as for MUTE functionality - you might well be better off by simply coding your system to send silence when you wanted mute rather than use this particular pin. The chip datasheet is pretty thin on what mute actually does - so unclear if it will do it in a nice way or not.

[grand lance]

ok. i got it to play, but it sounds very wierd

lotcs of rythmic clicking and sounds like a neighbor blasting it through the wall, clear enough to tell what it is, but not clear enough to enjoy it.

yeah. i cant fix it

in case its a coding issue, here it is https://pad.riseup.net/p/add_collor

[versed torrent]

Do you mean it is a noisy version of what you are expecting - like the right audio is there plus other junk.... or that it is all just junk?

[grand lance]

https://www.adafruit.com/product/987

[grand lance]

So, sleep didnt work, but it did stop it after 10 seconds so i dont have to deal with the choppy for long

[grand lance]

the feather keeps malfuntioning

[glacial spruce]

When I have situations like that, I usually eventually discover it's running out of resources (like a memory leak), corruption (buffer overflow, wild pointer), or a timing problem.

[grand lance]

i removed alot from my code to test just one component. i guess ill have to put alot of resourses on an sd card. currently its crashing and failing to show up unless i put it in to safe mode, which i only seem to be able to do by accident

ill figure it out later. i need sleep. i was unable to last night

[last prawn]

I can't tell for sure looking at your code - but I'm using that same i2s decoder and its really tricky to manage getting it samples while also using lots of i2c devices that fire off interrupts ( like your touchscreen ).

I'd try not enabling the touchscreen and see if it sounds better

[grand lance]

i removed everything and it keeps malfunctioning the board

i wish i didnt need this rasberry for class. i could use that.

[grand lance]

i have a metro, should i swap out for that?

[last prawn]

I'd verify what the cause is before doing anything with hardware (also switching out the controller may not be the right solution either way). If changed your code to only have the i2s device and play the sample does it still sound distorted?

[grand lance]

i dont know. the feather keeps crashing

[last prawn]

Hrmmm - I'd back up your code from the device, put it in bootloader mode, then reflash with circuit python

[random bone]

Playing audio while doing display operations will cause hiccups in the audio. The redisplay logic can take too long

I don't know why it's crashing: is it really crashing hard, or is it throwing Python excceptions?

[last prawn]

After that you can try out the examples on https://learn.adafruit.com/adafruit-i2s-stereo-decoder-uda1334a/circuitpython-wiring-test and see if it glitches still just to verify that something else isn't going on

[grand lance]

https://pad.riseup.net/p/audio-keep

this is what i am trying to run.

also, the sine wave runs perfectly

[last prawn]

Yeah - this is a thing. Theres a thread about audio DMA and I added partial refresh functionality to the display library for sd1306 for my project and it still mostly takes too long (its somewhat manageable if I sacrifice some responsiveness and use sufficient latency)

Next I'd add in the touch display - if that causes the glitching you're hearing then it might be tricky to have both at once

[random bone]

the sine wave is tiny and in memory, no computation or loading from a file requiredd

[grand lance]

i keep having to delete everything while in safe mode and somehow reset the board entirely then reinstall everything

[random bone]

you should not need to do that at all. Just press reset while in safe mode

[grand lance]

and it goes right back to acting like a usb device

[last prawn]

Thats true sometimes - but I have lots of issues with the boards on windows and a lot on my m1 mac (which isn't officially supported well for lots of things yet)

[random bone]

if the program continues to crash, don't run the program as code.py. Rename it to say, codex.py, and then import it from the REPL. Then you can test it without having to undo the auto-running of code.py

you don't have to have a code.py

i test all the time from the REPL

[grand lance]

right now the feather is connected but not showing up at all

[random bone]

you can only import it once. You have to ctrl-D to re-import it

is that because it was crashing?

[grand lance]

yes.

[random bone]

if you do a slow double-click, you should be able to get into safe mode, and then rename code.py

[grand lance]

like i said, the only sure fix is a full reset. i have to nuke it

[random bone]

you want the second click to happen while the neopixel is yellow. if you click reset during that time, you will go into safe mode

that is the way to enter safe mode manually

see the second paragraph here: https://learn.adafruit.com/welcome-to-circuitpython/troubleshooting#circuitpython-7-dot-0-0-and-later-2978455-21

[grand lance]

i altered the code name

[random bone]

my reaction time isn't good enough to see the yellow and then click, but a slow click will get the click during the yellow phase

[grand lance]

and the code isnt working at all

i saw some code for running other code but cant remember where. that may help

i dont understand this. my original code works just fine, but the trimmed down code crashes my feather

[last prawn]

Can you link me to the trimmed down code again?

[grand lance]

https://pad.riseup.net/p/audio-keep

[last prawn]

So what I'd do here is open the repl and type the lines 1 at a time to see which one is crashing it

[random bone]

or just add a bunch of print statements

are you seeing a backtrace, or is it hard-crashing to safe mode

[grand lance]

ok. found where it stops i did use the print statement

code.py output:
SPI Loaded
audio bus Loaded
Sine set
sine played
Sine test done
Traceback (most recent call last):
then it just flashes 2 red

i figured it out. my fault. i needed an r on the mp3 line. instead of reading the code past that, it seems to have locked it

[random bone]

what version of CircuitPython are you using?

it should not crash when trying to print the traceback

[grand lance]

most recent. unsure of actual numbers. i may even be using the beta

mp3 = audiomp3.MP3Decoder(open("Fields.mp3", "b"))

this was the trouble line

mp3 = audiomp3.MP3Decoder(open("Fields.mp3", "rb"))

this is what fixed it

[random bone]

the version should show up at the top of the REPL stuff. It is also in BOOT_OUT.TXT

we fixed some traceback printing problems not so long ago, so if you are using a beta, instead of 7.1.0 or 7.1.1, you may be using a version without the fix

but 7.2.0-alpha.1 should be ok

[grand lance]

it wasnt a traceback issue, it was human error

[random bone]

but Traceback (most recent call last): should have printed more. Did you just leave that out, or did it not print anything after that line?

[grand lance]

printed nothing. i left out the read capability of the mp3 line

[random bone]

but it should print something. It is a bug that it didn't print a backtrace. It should have indicated the error line. So that indicates an internal error. I just wanted to check the version to see if you were using a version before we fixed some traceback issues.

that is, it is not your fault that it printed nothing

[grand lance]

reguardles , that little r fixed everything

[random bone]

i understand, but if there is a bug, I would like to get it fixed.

it would have saved you time if it printed a backtrace and indicated a line number

[grand lance]

ok. like i said, nothing printed after that line. my feathers neopix immediately started blinking 2 slowish red and failed to work till i enterd safe mode and deleted the code file

[random bone]

could you tell me what is in BOOT_OUT.TXT?

in CIRCUITPY?

[grand lance]

Adafruit CircuitPython 7.1.0-beta.3 on 2021-12-13; Adafruit Feather RP2040 with rp2040
Board ID:adafruit_feather_rp2040

wow, those updated fast. i just got this board last month or 2

[random bone]

Thanks! It would be good to update to 7.1.1. We've fixed a number of bugs since beta.3, but beta.3 should have the traceback fixes already. I will try out the missing "r" in some test code.

https://circuitpython.org/board/adafruit_feather_rp2040/

[clever rain]

hi there ı want to make a hydophone and ı will find some frequency in underwater . I will use piezo transducer(SMC26D22H13111). ı will make a pre-amplifer(opamp) circuit and ı will amplifier piezo output signal(voltage), then I will make a band-pass filter my hydrophone will detecet between(20khz-50khz) ı think bandpass filter remove some noises and thruster(motor) vibration. I will connect this output STM32F4 DİSCOVERY microcontroller. Then ı want to use FFT and DSP ın my opinion ı can detect some voice and some signals. ıs it work is it possible.

can use Piezo transducer voltage directly to find voice and acustic signal
(Piezo Transducer [voltage]) -->Pre-amplifier-->Bandpass Filter-->ADC(stm32)-->FFT?
This is my pizeo Transducer

My main problem can ı use piezo transducer voltage directly to find acustic signal and voice.

there are a lot of parameter d33/d31/g33/g31 .this parameters how can effect my pre-amplifier or other circuit can you help me about this problem

[versed torrent]

@clever rain - If I understand, you want to put this underwater, and find interesting things to listen to and and then do some processing on.

The first question is what kind of signals are you going to detect with this? You give a bandpass filter range that is entirely above human hearing. You won't capture any voices humans make or any sounds we can hear....

You'd like to do FFT and DSP. STM32F4 is capable of doing that to a limited degree. It may be a very powerful microcontroller... but it's still a microcontroller! Depending on what signal processing you have in mind, it might be just fine, or it might be underpowered.

Tell us more about the signals you are hoping to capture, and what you'll use them for.

[odd oasis]

FFT for signals on an STM32F4 is fine for detecting frequencies and simple voice activity detection algorithms, but if you're expecting anything more advanced than a simple classifier, you'll probably need a separate microprocessor or PC to do the heavy lifting.

[last prawn]

It might be good to clarify what you mean by "detect" and "find" here - if you mean "Can I filter out other sounds so that I can record some voice/other sounds with a hydrophone" the answer is different than "Can I use the controller to classify what those sounds are?"

[clear cliff]

I build a breakout board for the Knowles SPH0641LU4H-1 Utrasonic (80Khz) PWM microphone. However, it seems like the PWM input code is hard-coded to only sample at a maximum of 22Khz, is this correct? What I am doing is benchmarking a capture, and calculating the time per sample, it seems no matter what samples rate or number of samples used, it works out to 22Khz every single time.

sample_size = 1024
def benchmark_capture():
    global sample_frequency, sample_size, time_per_sample, hz_per_sample
    start_capture_time = time.monotonic_ns()
    mic.record(samples, len(samples))
    sample_duration = time.monotonic_ns() - start_capture_time
    time_per_sample = sample_duration / sample_size
    sample_frequency = round((1 / time_per_sample) * 10**9)
    print(sample_size, "samples took",sample_duration,"ns or", time_per_sample, "ns per sample", "| Max Sample Frequency =", sample_frequency, "| hz_per_sample=", hz_per_sample)

# Main program
while True:
    left_overs = benchmark_capture()
    print ("left_overs=",left_overs)
    time.sleep(1)```

Output:
```1024 samples took 45898438 ns or 44822.7 ns per sample | Max Sample Frequency = 22310 | hz_per_sample= 85.9375
left_overs= None```
sample_size = 256
```256 samples took 12695313 ns or 49591.1 ns per sample | Max Sample Frequency = 20165 | hz_per_sample= 343.75
left_overs= None```

The sample rate and sample size seems to have little effect on how long the time per sample takes.

[solemn flint]

Can you clarify what MCU you're using and what microphone library? You say "the" PWM input code, but it's not obvious what that necessarily is.

[clear cliff]

Pi Pico, mic = audiobusio.PDMIn(board.GP0, board.GP1, sample_rate=sample_frequency, bit_depth=8, mono=True)

Regardless of the sample rate specified, every sample takes around 45776.4 ns to complete. Limiting the max sample rate to ~21.8Khz

Even settings a sample_rate of 0 results in the same thing. I wonder if it is just ignoring that entire variable on the PI Pico.

print (mic.sample_rate) gives a result of 44099

[random bone]

You are right, it is a bug. It does not change the sample rate. Let me check the code history.

[random bone]

Here is rebuild that just uses the passed in sample rate. mic.sample_rate will still be a little different. However, internally the PDMIn code uses a 44kHz windowed sinc filter, so I am not sure this is going to work for you. It also may not be able to keep up at high sampling rates.

[clear cliff]

Thanks, will give it a try a bit later. Do I just need to copy the uf2 file to the disk and it will update that library or do I need do compile something?

[random bone]

that UF2 is circuitpython, just copy it to RPI-RP2

[clear cliff]

I do get returns back faster now, but only once in a few reboots, it doesn't work a 2nd time unless I re-write 1's into the sample buffer. But even then it fails to return anything after a few cycles. I haven't yet checked to see if the buffer data is even valid.

sample_frequency = 22000
sample_size = 512

mic = audiobusio.PDMIn(board.GP0, board.GP1, sample_rate=sample_frequency, bit_depth=8, mono=True)
print ("mic.sample_rate=",mic.sample_rate)
time_per_sample = None

def benchmark_capture():
    global sample_frequency, sample_size, time_per_sample, hz_per_sample
    samples = array.array('B', [1] * sample_size)
    start_capture_time = time.monotonic_ns()
    left_overs = mic.record(samples, len(samples))
    if left_overs:print ("left_overs=",left_overs)
    sample_duration = time.monotonic_ns() - start_capture_time
    time_per_sample = sample_duration / sample_size
    if time_per_sample:
        sample_frequency = round((1 / time_per_sample) * 10**9)
        print(sample_size, "samples took",sample_duration,"ns or", time_per_sample, "ns per sample", "| Max Sample Frequency =", sample_frequency)
    else:
        print("sample_duration=",sample_duration)

# Main program
while True:
    benchmark_capture()
    time.sleep(5)```
output:

mic.sample_rate= 44099
512 samples took 976562 ns or 1907.35 ns per sample | Max Sample Frequency = 524288
512 samples took 976562 ns or 1907.35 ns per sample | Max Sample Frequency = 524288
512 samples took 976563 ns or 1907.35 ns per sample | Max Sample Frequency = 524288
512 samples took 976562 ns or 1907.35 ns per sample | Max Sample Frequency = 524288
512 samples took 976563 ns or 1907.35 ns per sample | Max Sample Frequency = 524288
512 samples took 976562 ns or 1907.35 ns per sample | Max Sample Frequency = 524288
sample_duration= 0
sample_duration= 0
sample_duration= 0```

The samples is still all 1's, so it is fast because it is not returning anything valid.

[random bone]

Hmm, The input sample_rate and the output rate should be close. I may not have fixed what I thought. In any case, could you open an issue, with sample code and output from your original tests? Thank you. https://github.com/adafruit/circuitpython/issues/new/choose

[clear cliff]

Summitted (went back to 7.1.1 to double check uploaded code) https://github.com/adafruit/circuitpython/issues/5914

[grand lance]

if i have a radio and an mp3 player in a project, where does the audio for the radio connect? to the decoder or to the amp?

[shut hazel]

Howdy! I have a noob project question... I've successfully figure out how to play music from the RP2040 feather and connect a speaker and amplifier. I would now like work on a separate project using the same components but adding a microphone to record short phrases that can play on button press. Will this work on the RP2040 feather? If so, what mic component can I use and how many recordings will the board hold? I'm not super worried about sound quality. Down the line... I would also like to work on a project that uses speech to text (probably from Google Cloud) on a connected board so using the same mic components would be great. If anyone can point me toward a project similar or documentation that would be most appreciated!

[random bone]

we have audiobusio.PDMIn, which can read short segments from a PDM microphone. However, it has to record to RAM, which is the limiting factor. Right now we don't have a bulk AnalogIn capability that reads analog signals. Both would be limited by RAM. Writing to an SD card or CIRCUITPY can have unexpected latencies, which are not so easy to handle. This is a piece of missing functionality that is not yet solved.

[shut hazel]

Gotcha! Thanks so much for the quick reply @random bone and for saving me hours of internet searches 😆

[glacial spruce]

If I'm reading your question correctly (and I may well not be), both the audio output of the radio and the audio output of the MP3 decoder would go to the amplifier, presumably with some sort of switch to select which one to listen to.

[grand lance]

kinda. the component that selects the device will be turning on the device while turning off the other devices. a switch will lock the radio or mp3 to be on reguardless, but both will not run at the same time. not sure how to do the last bit and i still think analog when there is a digital solution

if i use a latching switch mounted to a rotart encoder, i can press it at any time and switch between radio and mp3 as the background player, but still would have to use the rotary encoder to view its data

[glacial spruce]

I was thinking analog, while I'm dimly aware that a radio could provide some sort of data, I don't really think about that much.

[grand lance]

im just not sure if it needs the decoder or just the amp. i think its just the amp

[glacial spruce]

If what needs the decoder? I'm beginning to realize you and I may have different notions as to what's really going on here. I had thought you had three items: an MP3 decoder, a radio, and an (audio) amplifier.

[grand lance]

yes

the radio

[odd oasis]

Depends on the radio's output? If you're using a traditional RF module, there's nothing to decode. If the radio spits out some kind of I2S or other digital format, you'll need to decode something.

[clever rain]

sorry ı was working my final exam and ı see your answer so late 😦 . First of all ı want to make a hydrophone my piezo tranducer resonance frequency(43khz) because ı want to detect UBL(Undeerwater Locator Beacon) this Locater produce 45Khz signal(https://www.rjeint.com/wp-content/uploads/2020/04/ULB-362C-ULB-362C-PL-Beacon-Data-Sheet-1.pdf) I have ROV and ı want to detect that. so ı want to make a hydrophone.I read a lot of documantion about hydrophone design, there are a lot of important things for example transducer and circuit. My plain this UBL vibrate my piezo tranducer and this transducer will produce voltage but this voltage(signal) will be so low (10-20mV) first I will amplifer(pre-amp max 3.3V because ı dont want to saturate my ADC port ) voltage with opamp circuit later ı will read this Voltage(Signal) with ADC port. But this signal will be time domain ı will use FFT alghorithm and ı will convert this signal frequency domain after that ı can detect all frequency maybe I can use FIR filter adn Banspass filter ı think CMSIS library have this alghoritm and functions.

this is my ULB(UNDERWATER LOCATOR BEACON)'s datasheet:https://www.rjeint.com/wp-content/uploads/2020/04/ULB-362C-ULB-362C-PL-Beacon-Data-Sheet-1.pdf

this is my piezo transducer: https://www.steminc.com/PZT/en/piezo-ceramic-cylinder-26x22x13mm-43-khz

https://github.com/YetAnotherElectronicsChannel/STM32_FFT_Spectrum_Analysis

sorry ı cant speak english very well. I m trying to improve my english I m student. maybe I cant explain my idea

If ı can do that.I will produce my signal with transducer and ı make my DVL(Doppler Velocity Location). First ı will detect other signal if ı can do that ı will produce my signal and ı will detect this signal later ı will processing this signal.

[versed torrent]

Ah - okay - So - as I understand it - the beacons are doing something like giving a 10ms pulse every second at some ultrasonic frequency - like 45kHz -

So you want several times that sampling rate - let's say 200kHz..... if you want 1kHz resolution ---- well, you're going to need a 200 point FFT... so - let's say a 256 point FFT... and you'll need to do that at minimum (I'd think) of every 2 ms. (in order to get a few readings during a ping) - so.... every second, 500 256-point FFTs.

[clever rain]

this is turkish technofest competition ı will join this competition.

https://cdn.teknofest.org/media/upload/userFormUpload/TEKNOFEST_2022_İnsansız_Su_Altı_Sistemleri_Yarışma_-_Şartname_NnxRq.pdf

[clever rain]

I try this with MATLAB. It's working

this is not FFT this DFT.

[versed torrent]

So - we're talking more than about 5M operations per second. If you're doing them floating point (float ... not double!!) -- then I think the STM32F4 should be fast enough

Sure - but FFT is just an implementation of the DFT that is, well, fast! And for 256 points, the standard radix 2 based FFT is fine. You should be easily able to find the algorithm on-line.

Now - because your FFT would be only 256 samples ever 2ms... and there are 400 samples taken every 2ms... this whole enterprise should be pretty easy.... You could probably press to 512 point FFTs.... but you'd need to overlap your samples (not hard) - and it would increase to ~11M operations per second... still doable, and it would double your frequency resolution.

[clever rain]

thank you very much ı will try it ı will be here again.and ı will ask a lot of question 😄 thank for helping

thank you very much

🤩

[last prawn]

Noise might be an issue but you might also want to look in to using autocorrelation and staying in the time domain

Your original idea of just bandpassing around 43khz and then using a threshold to look for spikes might be accurate enough if that tone is that distinct and there isn't a lot of noise in that range

[bronze locust]

Anyone have recent experience with Mopidy and/or Pi MusicBox or similar? I primarily play music through a YouTube Music subscription and am looking into how feasible it is to set up a little home entertainment system for playing music without having to use my laptop or phone. I see Mopidy had support for Google Play Music (which was better that YT Music in every way), but not sure if there is a working YT Music integration yet.

aha, I see there's this
https://mopidy.com/ext/ytmusic/
which is not in the official Mopidy GitHub org.

looking at the star count and bug reports, not sure it's functional.

I guess an alternative would be to take a cheap commercial home entertainment system like a Roku stick and plug it into a little display

as far as I've seen, Roku (as well as Chromecast, etc.) expect you to have an HDMI display. What's the cheapest display I could use with one of those things?

ok I see that Google makes a product called Chromecast Audio that runs headless, and is controlled via the Chrome browser. Sells for $50

...and it has been discontinued?

would a cheap Raspberry Pi have enough juice to run a headless X server, load YouTube Music in a web browser and accept the occasional incoming VNC connection? (as annoying as that sounds)

[random bone]

I think so, it is designed to do video fast, and the rest is not that taxing

might depend on which model

[bronze locust]

Nice, thanks. I'm hoping to run it in a way that doesn't even use video. "YouTube Music" is the (worse) replacement service for the now-discontinued Google Play Music. They have a music library and also pull in results from YouTube videos, but I'm hoping there's a way to disable video playback from the latter in their browser interface.

[random bone]

i always wondered about that, for YouTube Music.

[bronze locust]

I'd expect the second-most taxing thing to be the fortress of JavaScript-based OS-within-browser that Google uses on the site.

[random bone]

"song mode": https://support.google.com/youtubemusic/answer/6313574?hl=en&co=GENIE.Platform%3DAndroid

[bronze locust]

there are articles for "Android" and "iPhone & iPad" - does it apply to the browser interface?

behold the Settings dialog in the browser interface

[random bone]

ha: https://support.google.com/youtubemusic/thread/79624347/how-can-i-switch-between-music-and-video-mode-listening-youtube-music-on-pc-browser?hl=en

https://www.reddit.com/r/YoutubeMusic/comments/l52stf/no_musiconly_mode_in_web_player/

[bronze locust]

curious why that's a "ha"?

[random bone]

it says "no"

[bronze locust]

I guess I'm just used to that

big tech companies with "support" teams that are 1000 miles from the people who have any influence on product development

if you beat the final boss of that dungeon, your prize is "Yes! We acknowledge that this is a wanted feature!"

(or "Yes, this is a bug! Congratulations, we agree with you!")

(no bug tracker link tho)

oh man, on the reddit there's this comment

Been wishing for this feature from a LONG time. Not to save bandwidth, but because many user created playlists contain videos and this feature will in theory switch to better quality songs.
that's great

[random bone]

I never understood the appeal of desktop youtube music if it's going to play videos. I use spotify desktop and I certainly wouldn't want videos consuming cycles in the background. But the reddit post seems to imply https://music.youtube.com might do the right thing??

[bronze locust]

not that I can see. I'm a premium subscriber and I do not see any option to disable video

what I really should be doing is looking for how to disable all video playback in Firefox

[random bone]

https://www.google.com/search?q=youtube+music+audio+only

further down the page there are some more interesting results

[bronze locust]

"YouTube Music's Free Tier Is Becoming Audio-Only" - wow, a free feature that is not available in premium. welp

the chrome extension sounds roughly equivalent to disabling video playback in the browser

I fear that the actual stream traveling over the wire may be a video + audio stream, so that even if the browser doesn't play the video, it's still being wastefully sent

and, more to my issue, if that's the case, then I'm guessing Firefox (and most web browsers) don't have an option to "play this media, but just the audio, not the video"

oh, looks like if you navigate away from the playback page (e.g., to the homepage or some track listing or whatever), the video goes away

or you can remove these classes from the song-video div

https://tenor.com/view/why-not-both-why-not-take-both-gif-11478682

it's nice that I can test out this solution (user script to disable the video player in CSS, so that I can't even get video by accident) on my laptop and then move to Pi once it works

[bronze locust]

update: Stylus extension with this user stylesheet works great. It even still shows actual album covers (images) for non-videos.

#song-video {
    display: none;
}

I'm now looking into getting this Bonnet to use with my Pi Zero 2 W https://www.adafruit.com/product/4037

[bronze locust]

this indeed seems to be a pretty significant obstacle, on a Pi Zero 2 W

[bronze locust]

potential-nevermind: Chromium significantly outperforms Firefox on this machine.

[bronze locust]

update: mopidy with the youtube extension can access YT Music! This seems much better than the VNC solution in every way.

[clever rain]

how can make this signal with matlab (code)

[odd oasis]

So, that's a 45kHz pulse in a 1 sec period?

[clever rain]

yes

[odd oasis]

So, something like y = square(2 * pi * t, 100/45000);?

[clever rain]

signal is 45 khz

[odd oasis]

Oh, sine wave in a 1% square envelope.

Hmmmm

[clever rain]

yess

[clever rain]

ı want to 45khz not 37.5 khz

[odd oasis]

Hoo boy my MATLAB is rusty

[dusty rivet]

Same, and I was just using it back in November lol

[odd oasis]

Fs = 1e6;
t = 0:1/Fs:3;

pulsewidth = 2e-2;
pulseperiods = [0:10]*100e-2;

x = pulstran(t,pulseperiods,@rectpuls,pulsewidth) .* sin(2*t*pi*45000);

plot(t,x)
axis([0 0.1 -0.5 1.5])

Sheesh that took way longer than it should have

It's nice that I can play with MATLAB scripts online in-browser now.

[toxic tartan]

ok, stupid question, but im using the EMG kit, and i'm running it connected to my computer. i dont have a usb isolator, but could i use a usb cable with the data lines connected to my computer and the power lines connected to a usb battery bank?

[solemn flint]

That's probably a bad idea. Even though the USB data lines are differential, they'd still want to have a common ground reference coming through the cable with them.

[nocturne stone]

Hello! Could I use this module if I only have access to one pin for left & one pin for right audio channels?

http://www.tianjiarun.com/zb_users/upload/2020/04/SJR-BTM525_SPEC.pdf

[glacial spruce]

I don't understand the question. For input, or output? What kind of access?

[nocturne stone]

Hey @glacial spruce, so the datasheet has those SPK pins which I thought were for audio, but I guess that is actually OUTPUT for when the chip is used as a receiver. Would the AIO pins be what's used for audio?

I'm trying to build a PCB for this to add a bluetooth audio transmitter to a device. I was also looking at it because it would be nice if I could add a mic input as part of the bluetooth.

I'm wanting to tap into the left/right audio channels that are connected to an audio jack. I may follow something like this, since I know where to solder the pins to, but having mic input AND BT 5.0 would be a big improvement.

https://youtu.be/v8Kp8cTOKQo

[glacial spruce]

I'm not seeing any AIO pins, but I do see some MIC pins

[nocturne stone]

It's there, it's the LEDs too

[glacial spruce]

Ah, those look like analog GPIO pins. They're likely not particularly good for audio use, but I suppose it's a possibility.

[nocturne stone]

After reading through this, it seems better?

https://fccid.io/2AMWOFSC-BT806/User-Manual/User-manual-4990883.pdf

There's also an integrated antenna, which would be better. I suspect the other one I was looking at was more focused on receiving

[glacial spruce]

I don't know, this one also offers mic inputs and speaker outputs.

[nocturne stone]

I'm not positive, but I think both are ok. The UGREEN Switch transmitter uses the CSR8670, which is the same bluetooth chip as the last image I sent.

https://www.aliexpress.com/item/4001193350465.html

I've seen the transmitter on Amazon, and it has good reviews

[clever rain]

thank you very much that's work ,but something is wrong first pulse period(10ms) is true but other pulse period(20ms) is wrong

[clever rain]

that is first pulse

that is second and other pulse

[odd oasis]

Ah, rectpuls is centered around x=0, so some displacement will be needed to shift the negative portion of the first pulse to get the first pulse right. Otherwise, you can just change pulsewidth from 2e-2 to 1e-2.

[clever rain]

I chanced but first pulse 0.05ms other pulses 0.1s

first pulse

second pulse

[clever rain]

ıt s have to start 1 not 0.995

[odd oasis]

Yes, the pulse is centered at x=0. You will have to shift the entire time to compensate for the negative half of the first pulse.

[humble whale]

Hi friends. First time posting in the Adafruit Discord! I have the PAM8302 amp board. Unfortunately I'm experiencing quite a bit of distortion. I'm powering it from 5V DC with a 8ohm speaker. Should the SHDN in be left floating? THanks for any help!

[lucid anchor]

Welcome! I am not super up on audio, so I'm probably not much help, but a quick look at a project that uses it has the SD pin not connected to anything, so I think that's ok. My only thought is have you tried tweaking the volume pot on the board? I'm wondering if, like most anything audio-related, when you have the volume too high, it distorts. Worth checking if you haven't already. But that's all I have to suggest! Someone else might have better troubleshooting ideas. But be patient - the folks who reply most on here are community members, so they answer when they can.

[humble whale]

Thank you for the quick reply and help. Yes, I've tried adjustng on both ends (output of the device feeing it and also nout attentuation trim onthe board) yet unfortunately still distortion at any volume level. Darn.

[lucid anchor]

We don't encourage crossposting, but you might get better help in #help-with-projects, simply because more folks hang out in that channel than this one. I suggest reposting your question there the next time you're around.

[rapid merlin]

I remember in using a speaker as a microphone for some high school physics experiment, is that possible in CircuitPython? I don’t see an PWMAudioIn class lol

[odd oasis]

No, that’s not possible, unfortunately. PWM audio works due to its ability to simulate analog signals over digital outputs. The inverse, however, doesn’t work that way. Most, if not all, microcontrollers can’t reliably receive analog signals on a digital input due to a gray area voltage range between logic high and logic low. You’ll need to use an analog pin to receive audio signals.

[rapid merlin]

Neat, the RP2040 feather’s analog pins can also do digital, might have to play with that

[inland matrix]

That works for almost every analog pin on a microcontroller, with the notable exception of A6 and A7 on SMD 328-based Arduinos like the Nano and Pro Mini

[odd oasis]

If you're thinking about attempting to output and input audio on the same pin, you may be able to, but definitely not simultaneously. The ADC and digital output peripherals are typically separate, meaning you can't run both on one pin at the same time.

You might also need separate external circuitry to amplify the input to voltages readable by the RP2040 ADC...

[bronze locust]

Is the power draw of a powered speaker negligible when it's getting zero line-in signal? Or is there a noticeable idling overhead (or is "idling" the wrong way to think about it)

[random bone]

It depends on the kind of amplifier that's being used to drive the speaker. A class A amplifier has a high quiescent current. But most powered speakers are not class A, they're probably class AB or D. https://www.quora.com/Do-class-D-amplifiers-use-less-quiescent-current-than-Class-A-B-amplifiers

[bronze locust]

Thanks @random bone! This also gives me the right terms to look further into it.

[versed torrent]

@bronze locust one thing to be aware of is that this is only true if the signal is truly zero V. I have seen some people send signals that weren't AC coupled to a powered speaker not realizing thier silence had a small DC component... Causing the speaker to use power continuously.

[bronze locust]

Ah, thanks. Would the headphone jack of my laptop with nothing playing be zero? (It's not just floating, the speaker is actually quiet)

[versed torrent]

Yeah - headphone outs of commercial equipment will be "AC-coupled" and be outputting zero volts at silence. You only need to worry about things like modular synths, and DIY audio circuits.

[bronze locust]

nice. And for the Adafruit RPi DAC board https://learn.adafruit.com/assets/66513, I guess the capacitors C4 and C6 (region B3 in the schematic) mean its output is AC coupled as well?

[dull basalt]

If you fail to plug in a load, but do plug in a 1/8" stereo plug to a modern headphone jack .. say for a television .. and have the volume up, does it risk damage to the TV?

(I use a 1/8" stereo splitter as a cheap splice to connect headphones to the end of an audio cable that runs to the heaphone jack of the TV and stays plugged into the TV 24/7 but I disconnect the far end where the headphones plug in).

[solemn flint]

That sounds okay to me. It would generally behave as the same sort of open circuit as not having the cable plugged in at all.

[dull basalt]

Thanks @solemn flint - I try to keep it loaded but convenience makes that less than likely at times. ;)

as an aside I get monster RF hash off my PC's headphone jack when I plug a (dual) RCA to stereo adaptor cable into it.

whoops I meant input jack (mic jack).

I have a nice shortwave set with stereo RCA output jacks.

I get 9 pounds of noise on the S-meter when that cable is connected to the PC.

(eton satellit recent model)

[covert portal]

I have this adafruit mic amp thingy, which is alright but it picks up a lot of background noise. I've heard I can desolder the onboard mic to attach it to a 3.5mm jack but does anyone know which parts I need to connect? I assume it's the two pins but I don't know which is which

[odd oasis]

It varies by device. A TRS stereo microphone, for example, would likely have Tip+Ring1 on one audio channel, Ring2 on the other, and Sleeve to ground, while a TRRS headset would use tip and ring 1 for left and right channels, microphone to sleeve, and common ground on ring2. There is surprisingly little standard available for such a common plug, part of the reason why you often see like 6 of these plugs on the back of old PC sound cards...

[covert portal]

I can probably trial and error which spots to use on the 3.5mm jack end - I was more asking about the pins on the mic amp thing https://shop.pimoroni.com/products/adafruit-electret-microphone-amplifier

[odd oasis]

Oh, the microphone is soldered in on the mic and amp through-hole pins?

[covert portal]

I believe so, yeah

my guess would be mic > microphone ring, amp > ground but it's just a guess

[odd oasis]

My guess would have been the reverse, but I haven't looked at it yet...

[covert portal]

also, just checked the manual that came with the mic, so I know the standard of the rings

let me see if I can trial and error it by tapping wires together

[odd oasis]

It might not even make a difference, if I'm thinking about it now...

[covert portal]

hmm, doesn't seem to pick up

the mic is CTIA standard so I did sleeve > mic and ring2 > ground but it didn't seem to make a difference

I was just touching the wires to the solder points though

[odd oasis]

Have you soldered the TRRS breakout ends of the wires?

[covert portal]

not yet, I wanted to check it was working first

[odd oasis]

FYI, you'll get better results with your connections if you solder one end. You need pressure to maintain a connection otherwise, and trying to do so on four unsoldered ends is far more difficult than two.

[covert portal]

I figured/hoped this would be enough to tell if it was working

[odd oasis]

It doesn't matter which end you choose to solder first, but I strongly recommend you choose one and test that way.

Wires are easy enough to desolder from my experience, if you have a solder sucker.

[covert portal]

Yeah, just more of a pain if it doesn't work lol

I guess I'm pretty confident that the leads are correct on the TRSS...

[odd oasis]

Better that than indeterminate test results though haha

[covert portal]

Let's see if it makes a difference

welp, you were totally right - worked first time

[azure panther]

I have an MP3 with a bitrate of 192 kb/s, which will not play on my pi. I've been advised to try lowering it to 128 or 98 kb/s. Can it be raised back up to 192 kb/s if these two lower numbers don't work?

[odd oasis]

Not without losses, from my understanding. It would be best to save a copy before making experimental modifications.

[azure panther]

Cool thanks. This pi isn't playing known good MP3s so I'm going to try another pi

Is there a low-overhead website that will allow me to play sounds? I want to make sure my audio port is working first

[dull basalt]

I use a local application (audacity). Pretty sure it's ported to other systems besides Linux.

Should be able to write a file at other sampling rates as well, though I don't remember what it'd do if it was sourcing a higher bitrate.

[azure panther]

I meant more that would have files I can play, Audacity from my experience is for editing files you already have. The files I have aren't working

[dull basalt]

i installed enough audio tools in linux to have some samples provided with those tools (in particular, to test left, right and combined L+R speakers ;)

archive.org often offers ogg vorbis files which I think are audio-only or maybe they're also used in video with a soundtrack.

Some artists used to publish directly to archive.org.

You might find Grateful Dead, John Popper (Blues Traveler) and also Dave Matthews there.

https://archive.org/details/BluesTraveler

[azure panther]

I think something's wrong with my audio jack. I dropped something metal and shorted some contacts nearby a few days ago. I'm going to try a fresh pi.

Nope, not my pi. Let me test the speakers

Speakers work. So it's some kind of software issue

[dull basalt]

For desktop amd64 linux I like pasystray to adjust sound volume and select channels.

So I can choose a different sink and source.

Including my logitech headphones and the built in audio jacks on the desktop PC chassis.

[odd oasis]

Is your Pi configured to output audio over HDMI, perhaps?

[azure panther]

.....

You can't see but I'm making a "disappointed in myself" face rn

I always always always forget that that's the default

there it is

[lucid anchor]

Oh, seriously? That's the default? I haven't run into this yet, but I almost certainly would have.

[azure panther]

Yuuuuup. It's real annoying

[odd oasis]

I wasn't 100% sure, but I recall reading that somewhere as I setup my Pi 4...

[lucid anchor]

Oof.

[azure panther]

Interestingly for some reason now the two options are:

0 Headphones
1 MAI PCM i2s-hifi-0

Previously 1 was something about HDMI

[azure panther]

@hazy hound It wasn't the bitrate, it was the default setting of the pi being to output audio over HDMI!

[dull basalt]

'alsamixer' often works in Linux. Maybe not always.

run it in an xterm usually.

[azure panther]

how would I go about adding "empty" space to an .mp3? Like just no sound for a few seconds

using Audacity

ahh the key I was looking for was "silence" lol,

[lethal slate]

Enjoy it.

[trim steeple]

https://twitter.com/dcbriccetti/status/1495872261202616321

Dave Briccetti (@dcbriccetti)

“Has anybody been able to record and plot the 40khz signals from an ultrasonic rangefinder? Here in Audacity I changed the sample rate to 96khz, but the frequency analysis window is only showing up to 20khz. I also tried Sonic Visualizer. https://t.co/Y4y71EXIlI“

Twitter · Feb 21, 2022, 9:25 PM

[random bone]

see the project rate in the lower left: it's still only 44100 kHz

[trim steeple]

Hey, @random bone

[random bone]

I think you want to set the rate before you import the data?

[trim steeple]

I set the rate, then made the recording. Exported and loaded into Sonic Visualizer:

Also set to 96khz here:

[random bone]

you're recording an analog signal?

[last prawn]

Hrmmm, did you actually use your macbook microphone? It might not let you record above 20khz

[trim steeple]

Output from HC-SR04 into MacBook Pro’s mic near left speaker.

[random bone]

mic jack?

the circuitry or sw on that may be band limiting

[trim steeple]

I could try USB or the “mic in” analog jack.

[last prawn]

I wouldn't really trust that apple mic or line in to not severely bandlimit your signals.

[fickle leaf]

You will need a special audio interface for frequencies above 20kHz.

[trim steeple]

My goal is to visualize the 8 40khz pulses from the rangefinder, to make it more real for my students.

[random bone]

that may be easier done with an oscilloscope

[last prawn]

Based on your recording theres a brick wall aroudn 20k, and then it applied dither.

[trim steeple]

@random bone Oscilloscope with a mic plugged into it?

[random bone]

yes

or output signal from the rangefinder , if the mic is not sensitive to those freqs

[last prawn]

Finding a mic that has a good frequency response that high probably won't be easy either (you could try a cheap reference mic). I'd go straight out of the range finder

[trim steeple]

I’m hoping to see a square wave of the 8 cycles at 40khz, somewhere.

[random bone]

CGrover has a lot of audio experience

[trim steeple]

The range finder only would reveal the trigger and echo signals. (Not the 8 pulses.)

[last prawn]

If you wanted to use an audio interface you'll probably need something that goes up to 192k to guarantee the input filter isn't cutting out 40k. 96k is cutting it close ( 96k will want to filter anything out above 48k - but since filters aren't ideal it probably starts lower)

[trim steeple]

Well, good to know it’s not something easy that I’m missing. This may be more trouble than it’s worth.

[fickle leaf]

Even with a super audio interface, finding a capable microphone presents a challenge.

[trim steeple]

It can be enough to show the kids the realtime Audacity recording lighting up real big when there’s nothing they can hear. (As I hold the rangefinder’s speaker over the MacBook’s mic at the left speaker.)

[fickle leaf]

Can you tap into the rangefinder’s output amplifier with the scope’s probe?

[random bone]

would the mic from an ultrasonic rangefinder be fine? If you have another to sacrific?

this is one of those cheap rangefinders that everybody sells for Arduino, etc., right?

[trim steeple]

Oh, fantastic idea.

[trim steeple]

Yes

[random bone]

you could tap into the mic on the original rangefinder

[trim steeple]

Oh, and see the echoed pulses?

[random bone]

yes.

looking at one of these, it looks the identical transducer is used for both sending and receiving. That's not so surprising: you can use a speaker as a simple microphone

[trim steeple]

Thanks, all, for the great info and suggestions!

[random bone]

the terminals are accessible on the back.

If you have a dog whistle, you could demonstrate that too

[trim steeple]

Hah! I think I’ll order one.

[last prawn]

Wow... manufacturers still don't measure frequency response above 20k... thats frustrating. Most microphones will go above that and it doesn't really matter if they aren't flat if you know the response.

[fickle leaf]

There used to be some debate about the benefits of capturing ultrasound frequencies to recreate harmonics in the range of the ear, but not much became of that.

[last prawn]

Yeah, considering the snake oiliness of the industry is feels like not measuring up there is leaving money on the table so I'm just surprised they don't.

Measurement is an area where capturing high frequencies is useful (we get requests in the ghz range at the company I'm at). Also, we used to pitch a lot down when I did sound design and having that extended frequency range helped a lot.

[trim steeple]

Could this be the 8 pulses? (I’m new with Oscilloscopes, and I have a tiny one.) I probed the speaker connectors so these are outgoing.

[primal lagoon]

Take a look at Bat detectors. I did a lot of work many years ago on these (Before expensive commercial ones arrived). They can tell the calls of bats apart (Audible and visual). A cheap version would be adequate as you already know what signal you are looking for. We were using laptops and headphones to watch them.

[glacial spruce]

Sure looks like it to me

[clear cliff]

@trim steeple , I am using a 80KHz capable PDM microphone to detect and FFT an ultrasonic (~38KHz) mechanical remote on a Pi Pico. The SPH0641LU4H-1. I also made a breakout board for it, and it only cost me about $40 to get four boards and the parts.: https://oshpark.com/shared_projects/nfaNGKb1 (Schematic here: https://oshwlab.com/ZapWizard/radiation-king-radio_copy)

[median prawn]

Would this form a bandpass?

[versed torrent]

i don’t think so

[glacial spruce]

No, that's basically what's known as a "capacitive divider"

[median prawn]

Ok, I'll look into those

I had a feeling that if it did form a bandpass I would've seen it before

[tall mango]

What equipment would i need to get a line out from USB?

[solemn flint]

The obvious answer is a USB audio adapter, sometimes called a DAC. Are you looking for a particular chip suggestion?

[tall mango]

I tried a sabrent USB DAC, but it only has a headphone output and I'm not sure they're the same

[solemn flint]

They're not the same, but you can get a line-out signal from them in a pinch by adjusting the volume. You'd probably get higher quality from a dedicated line-out adapter, though.

[tall mango]

Yeah, I'm looking to make a cassette recording setup with an old cassette deck I got

[solemn flint]

Audio quality is probably not a huge concern, then, so that should work fine.

[nocturne stone]

Would I be able to use a TS3008 or a TS8675 to receive l/r audio signals AND transmit earbud MIC to the module?
https://www.tinyosshop.com/index.php?route=product/product&filter_name=ts3008&filter_description=true&filter_sub_category=true&product_id=1120
https://www.tinyosshop.com/index.php?route=product/product&filter_name=ts8675&filter_description=true&filter_sub_category=true&product_id=1121

I'm aware I can use the l/r audio signals on these, but I can't tell if it supports outputting the MIC signal

Would they go through the "SPK" lines?

[odd oasis]

I see MIC lines for microphone input, and SPK for audio output. Could you clarify what you’re trying to do?

[pearl depot]

Hi! Quick question, is it by any chance possible to use this featherwing https://www.adafruit.com/product/3436 to have direct access to the SD card as well? As in, on top of using it as a player, I'd like to read and write some data to that SD card. Haven't found any info about that yet.

[odd oasis]

The SD card shares the same SPI SCK, MISO, and MOSI pins with the VS1053, and has its own CS pin. All the pinouts are there in the learn guide, and you can learn more about SPI here: https://learn.adafruit.com/circuitpython-basics-i2c-and-spi/spi-devices
The code will be different if you’re not using circuitpython, of course, but the wiring and hardware concepts are all relevant regardless.

[pearl depot]

oooh right! I can then select who I'm talking to depending on what I'm sending through SPI. Was focusing on needing separate pins for some reason. Thank you! ^^

[boreal pilot]

Hey all, recently installed this speaker in my system and it works great. However, there’s a slight buzz before it plays the first sound. Any way I can fix this?

I’ve ran a 104 capacitor between the audio put pin from the arduino and the audio in from the amp. Didn’t seem to do much

[glacial spruce]

I'm unclear as to where you installed the capacitors, but it could be that it's picking up noise before there's an audio signal. Maybe a 10kΩ resistor from the audio input of the amplifier to ground to absorb stray signals could help.

Alternatively, you could control the power/enable signal to the audio amplifier and only activate it when the rest of the system is ready.

[boreal pilot]

Those arent bad ideas, ill give them a shot, thanks!

The capacitor is installed between my audio outpin pin (A0) to the L+ input on the amplifier

[glacial spruce]

So more of a high-pass/coupling capacitor than a filter.

[dull basalt]

What is analog filtering?

[deft silo]

it's just filtering of an audio signal without using digital means as i understand it

e.g. using an op amp with a negative feedback loop

oh! anyone got any opinions on this module

https://thepihut.com/products/adafruit-stemma-speaker-plug-and-play-audio-amplifier-ada3885?variant=30679065985086

not expecting audiophile sound or anything

just, would like a basic speaker

[haughty zenith]

I'm too lazy to do anything about it, but I suddenly imagine with the right software a pi pico connected to an sdcard could make an awesome mp3 player

[boreal pilot]

Any way to prevent my arduino from making a clicking sound everytime a sound starts and stops playing?

[boreal pilot]

(fixed the audio loudness issue)

[weary gyro]

It depends on the audio library you are using. Clicking from audio PWM outputs (like that from standard Arduino) is usually caused by the audio PWM driver being stopped and started, instead of continuously running. When PWM stops, the output voltage goes to zero, but for PWM the output should continue but at 50% duty cycle. One way to solve this is to "play silence". Another is to do something like analogWrite(audioPin, 512) immediately after stopping sound, but you may still get clicks

[boreal pilot]

I think the "play silence" solution might work best for me, i can just run it at the beginning

[weary gyro]

Which audio library are you using?

[boreal pilot]

audiomp3

its a circuitpython library

[weary gyro]

oh so not Arduino

[boreal pilot]

I suppose not

its a Feather RP2040.

so i guess closer to a raspberry pi than an arduino, lol.

[weary gyro]

No, you're right the first time. It's more like an Arduino, but "Arduino" and "CircuitPython" are two very different ways of programming Arduino-class boards

[boreal pilot]

I see.

[weary gyro]

In the CircuitPython case, I would recommend using AudioMixer instead of playing the MP3 directly with AudioOut.play()

Let me find an example

[boreal pilot]

Sweet, thanks!

[weary gyro]

# instead of doing something like this:
data = open("test.mp3", "rb")
mp3 = audiomp3.MP3Decoder(data)
audio = audioio.AudioOut(board.A0)
audio.play(mp3)
while audio.playing:
  pass

# try this instead:
mixer = audiomixer.Mixer(voice_count=1, sample_rate=22050, channel_count=1,
                         bits_per_sample=16, samples_signed=True)
audio.play(mixer) # attach mixer to audio playback
data = open("test.mp3", "rb")
mp3 = audiomp3.MP3Decoder(data)
# tell first mixer voice to play mp3
mixer.voice[0].play(mp3)

I assume your code looks like the top part. When the audio file is finished, CircuitPython shuts down the AudioOut. In the second part, an AudioMixer is created and attached to the AudioOut. This keeps things running, and then you tell the different mixer voices to play

[boreal pilot]

Interesting, thats awesome. Thanks for letting me know!

Now I just have to fix my VERY janky speaker. I nudge it a little bit and it shorts something out and makes my board reset.

can I ask you about that @weary gyro ?

[weary gyro]

I'd have to know what your audio output filter circuitry looks like. Speakers can like a short-circuit to microcontrollers, so that could be the problem. I would recommend feeding to a little powered speaker (like an old bluetooth speaker) or maybe even headphones. And I recommend using a circuit like this: https://github.com/todbot/circuitpython-tricks/blob/main/larger-tricks/docs/breakbeat_sampleplayer_wiring.png

(that circuit is for this demo: https://twitter.com/todbot/status/1494899839871512577)

[boreal pilot]

Unfortunately I’m limited to having a speaker in the unit.

it's an interactive toy thing

it's super super janky.

[weary gyro]

oh! also another thing. PWM Audio on CircuitPython can crash the board if you are writing to the CIRCUITPY drive while the program is running. I recommend stopping your program with Ctrl-C in the REPL before saving code.py or writing MP3 files

[boreal pilot]

Interesting okay

[weary gyro]

oh so you're using an amp, so I think your audio section is okay. I suspect it's the CircuitPython PWM Audio crashing problem.

[boreal pilot]

interesting.

[weary gyro]

I've experienced this problem so much I opened an issue https://github.com/adafruit/circuitpython/issues/6005

[boreal pilot]

i thought it was always my janky wiring

[weary gyro]

hey I've seen/done worse 🙂

[boreal pilot]

I was told from someone here to add a 10k ohm resistor from the input on the amp to ground as a bit of a filter.

might be picking up some outside interference

[weary gyro]

you need an RC (resistor/capcitor) filter on the digital PWM output of the RP2040 chip to turn it into an analog signal. I don't think it would be picking up interference. If you had nothing, you'd probably not hear anything bad. Your little amp will probably filter it fine

[boreal pilot]

even though there's a buzzing at the beginning?

[weary gyro]

Not sure what that is. I'd need to hear it.
If it's a high-pitched arhythmic sort of harsh noise that happens just at the start, that's your computer bogging down the RP2040 accessing the CIRCUITPY drive.
If it's a regular high-pitched noise, it could be power supply noise from whatever power supply you have driving your board or your PC's power supply if it's connected via USB.
In both cases, try driving your board entirely from battery, no computer, no AC power

[weary gyro]

That sounds like a capacitor charging up. Do you have a wiring diagram of how you have the RP2040 board hooked up to that little amp?

And does that noise happen every time you play a sound after the code starts running? e.g. if you play a sound -> stop -> play a sound in code?

[boreal pilot]

thats a 104 capacitor

No, it only buzzes after startup. every other time works fine.

[odd oasis]

If you have a capacitor charging up at startup, one possible workaround is to wire the SHDN pin to a GPIO and have it grounded for a short delay to allow the capacitor time to charge, before sending logic high to enable the amp?

[boreal pilot]

To be honest I’m not sure why the capacitor is in there. It was just recommended by a professor

[tall mango]

Is an "Audio Distribution Block" just a bunch of RCA jacks in parallel?

In that case could i just make my own with some RCA plugs?

[lethal slate]

import asyncio
import board
import time
import usb_midi
import adafruit_midi
from adafruit_midi.note_on import NoteOn
from adafruit_midi.note_off import NoteOff
from neopixel import NeoPixel
from adafruit_seesaw.seesaw import Seesaw
from adafruit_seesaw.analoginput import AnalogInput

def bpm_to_t(rate):
    return 60/rate/4.0

class Ticker:
    def __init__(self):
        self.bpm = 20
        self.tick = bpm_to_t(self.bpm)
        self.note = 38
        self.running = True
        self.tock = 0

async def loop(ticker):
    midi = adafruit_midi.MIDI(midi_in=usb_midi.ports[0],
        midi_out=usb_midi.ports[1],
        in_channel=1,
        out_channel=1)
    while ticker.running:
        midi.send(NoteOff(ticker.note, 0))
        midi.send(NoteOn(ticker.note, 120))
        await asyncio.sleep(ticker.tick)
async def read_ain(ticker):
    seesaw = Seesaw(board.I2C())
    a7 = AnalogInput(seesaw, 7)
    while True:
        raw = a7.value
        ticker.bpm = int(raw/1023*120)+20
        ticker.tick = bpm_to_t(ticker.bpm)
        await asyncio.sleep(0.1)

async def main():
    ticker = Ticker()
    await asyncio.gather(
        loop(ticker),
        read_ain(ticker)
    )
asyncio.run(main())

MIDI Click Track

[fickle leaf]

That amplifier already has DC blocking capacitors on the input pins on-board, so the external series cap on L+ isn’t necessary. A 10k resistor across L- and L+ could help; pin A0 is in a high-impedance state when the board first starts up so the amp inputs will likely pick up some random CPU noise during initialization.

Connecting the SDWN pin to a GPIO pin has a similar timing issue. The state of the GPIO pin is high during startup and won’t squelch the amp until after startup. The GPIO pin is actually in a high impedance state like the A0 pin. It just looks like a logic high because of the amp’s internal pull up resistor on the SDWN pin.

A capacitor on the SDWN pin to ground might help, but a 10uf cap would only provide a few milliseconds of squelching after power is supplied. Might be worth an experiment if the 10k resistor doesn’t do the trick.

[boreal pilot]

this gives me "audiomixer" is not defined

[weary gyro]

sorry that was a code fragment, you need to import audiomixer at the top of your file

[boreal pilot]

Thanks! is that a library I need to find myself?

or is it built into the board

nvm all good

it's pre built into my board

[weary gyro]

Then I think the startup buzzing is just something you'll have to deal with because of the particulars of doing PWM audio with that little amp board

[boreal pilot]

Ah okay

[weary gyro]

If you can experiment, you may want to try powering the amp from the "USB" (5V) pin instead of the 3V pin. That amp can be driven from 2.7V- 5.5V and maybe the noise is coming from the Pico. Or if your project is battery powered, run the amp directly off the battery. Chasing noise is a common pastime in audio projects

[boreal pilot]

I'm using a voltage step-up for the speaker

Now i have to find out why my speaker is constantly crashing. It's really strange

sometimes it crashes so hard it goes into safe mode, apparently.

[weary gyro]

not sure what you mean by the speaker crashing

but PWM audio in CircuitPython can cause CircuitPython to crash some times, especially if you're reading/writing to code.py or other CIRCUITPY drive files

[boreal pilot]

does this dumb thing and it's really bothering me, lol.

[boreal pilot]

would anyone know why this is happening? it's definatley stumping me lol.

[glacial spruce]

No, we can't know, we can only guess. I'm guessing a data format problem, a power/ground interference issue, or clock/data issues.

[boreal pilot]

Ah, that’s fair.

How can I begin to diagnose?

[odd oasis]

Easiest way is to hook up your output to an oscilloscope, if you have access to one?

Anything outside of that is basically careful guesswork...

[boreal pilot]

I believe we do. At the university

Might be able to look into it tomorrow

[boreal pilot]

Would it have something to do with the quality of this amp?

When I have no speaker in it works perfectly. Nothing wrong

[glacial spruce]

How can you tell with no speaker?

[boreal pilot]

However I feel as though I’ve damaged the chip itself

[boreal pilot]

The code works and it doesn’t reset and crash

[glacial spruce]

Note that the soldering on those pins looks iffy

[boreal pilot]

Yeah I just desoldered it

That blotch in the upper corner is leading me to suspect that something is wrong with it

[boreal pilot]

Removed my amp from the board and discovered an awful soldering job. Guaranteed something was bridging

[boreal pilot]

https://learn.adafruit.com/stereo-3-7w-class-d-audio-amplifier/downloads is telling me not to use this as a pre-amp

is this my downfall?

[fickle leaf]

It’s specifically designed for outputting to speakers rather than boosting a signal as an input to another amp. That’s why the guide says it can’t be used as pre-amp. Your first video shows the output of the L side going to a speaker as it should.

[glacial spruce]

Yes, a power amp produces a speaker level signal and a preamp produces a line level signal. Worse, a class D power amp produces a horrible pulse wave that only averages to the desired signal. A speaker filters this out due to its inductance and mechanical limits, so the high frequency garbage isn't a problem. But putting that signal into an input will make a hash of it.

[fickle leaf]

However some of the soldering of the input, power, and ground pins looks iffy, particularly the second ground pin.

[glacial spruce]

When my fridge died, I was disinclined to shell out $90 for a new CPU board, especially when it was obvious the original one had some bad solder joints and a burnt resistor. I didn't have a 2 ohm 2 watt resistor, so I made one out of eight 1 ohm 1/2 watt resistors. I bought an appropriate replacement resistor (uprated to 5 watts) but it ran just fine for years with the bodged one.

[boreal pilot]

I think I found my issue.

It was actually software side.

it still kinda janks out which im not sure why

But it's significantly better

So it will play once, jank tf out, crash, go again and be completely fine

Would anyone be so kind as to take a look at my code? Theres something i HAVE to be doing wrong.

[dull basalt]

hey so I am using the PAM8302 amp board connected to a 4ohm 3watt speaker. The issue is that I don't know if my board is shorted, or maybe this version of the board is notorious with having similar problems to this, or if this is normal lol, but I have checked over and over again from a variety of sources to make sure my wiring was all correct, and I came down to one thing. I have no idea if this is a problem or not, but the positive and negative outputs both replicate each others voltage, if I just have the wires their own connected to the speaker, the resulting voltage across both of them is of similar proportions to my vdc input, however if i directly connect a ground cable to the negative audio output, both of them share ground. I will upload a picture of my circuitry if needed tommorow, thanks in advance for anyone who might know anything about the subject.

[harsh gust]

Hi, guys. I'm kinda new to this community and I hope this channel is the right place to throw this question.

I'm really a newbie when it comes to building circuits and make things in general. I'm kinda crashing learning all the knowledge I need to know (it's pretty tough)

Anyways, I am trying to build a device that can detect ants stridulation with using piezo mic.

So I need an amplifier for that.

I found 3 different amplifier circuit so far.
But I don't know exactly what their difference is, and which one would be the best one for my use.

opt 1 : https://makingcircuits.com/blog/diy-contact-mic-circuit/
opt 2: https://breaksberlin.wordpress.com/2015/11/07/diy-piezo-preamp-with-hi-pass-filter/
opt 3: http://www.richardmudhar.com/blog/piezo-contact-microphone-hi-z-amplifier-low-noise-version/

I would be so much appreciated if anybody can teach me the difference, or give me any advice.

Thank you.

[dull basalt]

I recommend just getting something similar to the MAX9814 (make sure to get the one with no mic pre-soldered, since you already have one. The options you have sent here appear to be attempting to make an amp from scratch, which is probably not what you want, as it will come to worse and more expensive than if you buy a mic amp from the internet (such as the MAX9814)

[harsh gust]

thank you for your comment!
I'm taking a class that is focused on building circuits so I want to at least give it a try. And I already bought couple of components already..

Why building one of those circuit could be worse? and what is the difference with max9814?

I am curious. I will consider purchasing max9814 but I would like to know the difference, and at least try making those circuits.

[glacial spruce]

It doesn't have to be worse, but mic amplifiers are working with low level signals, so they're susceptible to noise, interference, and feedback. However, if you're looking at building circuits, rolling your own mic amp can be quite instructive.

[harsh gust]

thank you for your reply! Then what would be the difference between those circuits? And do you also think that using max9814 is the best option?

[glacial spruce]

Between which circuits? As with all engineering, what is "best" depends on what your parameters are.

My version of "best" often boils down to "what can I build without parts I have to wait for"

[dull basalt]

adding onto what madbodger said, smaller boards are typically preferred since a consumer-produced circuit with low-grade tools will likely end up in the current travelling a further distance, causing delays, electricity has speed too, obviously this delay won't be too noticable in an amp, but if you want to do real-time manipulation, it might cause some challenges

[harsh gust]

ohh I got what you mean. Thank you so much for your insights.

[harsh gust]

oh, ok so, since my purpose is to record ants sound and it is quite meticulous and sophisticated work,

I thought I should come up with an amp that has more gain boost and filter/noise cancelling feature

[harsh gust]

I was curious about the difference between option 1,2,3 I mentioned above.

[glacial spruce]

Option 1 is very basic, and will not provide much gain/sensitivity.

[harsh gust]

At this point, not in a sense of which one would be the best, but literally, what would be the difference

oh ok

[glacial spruce]

Option 2 gives a good bit more gain (up to 50dB), and the ability to tune the filtering. However, with that much gain, you will need to use good construction techniques, power supply filtering, and shielding to keep it from boosting certain frequencies unevenly or even oscillating. It's a decent design, but will require some engineering to work optimally.

Option 3 is a little simpler, and will give less gain.

The MAX9814 can offer the most gain (60dB) if you crank it all the way up. It's designed for an electric microphone instead of a piezo, but the impedance is high in both cases, so it should present much of a problem. You could, if you wanted, try adding a buffer (like option 1) between a piezo sensor and the electret chip, which would give a little extra bonus gain, and some impedance buffering.

For your use case, high gain and low noise are probably going to be your dominant parameters.

You might look at some high performance precision op-amps if you want to try rolling your own (the OPA211 might be a reasonable choice, it's extremely low noise and not horribly expensive)

[harsh gust]

ohh wow

[glacial spruce]

The OPA1611 might be a good choice too, and has a more attractive price. The various chip vendors offer lots of helpful information like this handy guide on op-amp noise: https://www.ti.com/lit/an/sboa345/sboa345.pdf

[harsh gust]

wow thank you so much for your help.

First I have found this guy on bandcamp

[glacial spruce]

Your use case is a little unusual, but intriguing

[harsh gust]

https://kuaishen.bandcamp.com/album/stridulation-amplified-compositions-with-the-stridulatory-organ-of-atta-cephalotes

he is an artist making use of ants sounds he recorded and he offers a simple circuitry that he insists he used for the recording

[glacial spruce]

That is bizarre and cool. I'm amused that he offers the schematics of his circuitry with the album

[glacial spruce]

LM386? That is a truly peculiar choice

[harsh gust]

this was the one and my classmates said it would not be a suitable choice

yeah that was what my collegues said too

and since I am such a newbie of these electronics stuff

I asked them and they suggested me those 3 options

[glacial spruce]

We all start out as beginners. The LM386 is a low power speaker amplifier, it's not really designed as a pre-amp. It would basically sort of work, but there are better choices out there.

Time for me to head out to dinner.

[harsh gust]

thank you so much for your insight. Have good one!

[harsh gust]

I just made a simple piezo-amp circuit and connected to my computer

but that is the result. And it keeps producing that noise no matter there is a 9V batteries or not.

And whenever I touch the piezo, it goes like that like a lightsaber

https://www.youtube.com/watch?v=aOJuCYgmPPE

this is the one that I followed. I'm not trying to use that but just tried building one to see how it sounds like.

[glacial spruce]

That's hum pickup from AC wiring in the area

[harsh gust]

what is ac wiring? you mean the 1/4' female component?

[glacial spruce]

No, I mean mains wiring, like to wall outlets

You probably will want shielded wiring to connectors like that, however.

[harsh gust]

ohh which parts should I use shielded wiring?

[harsh gust]

so you're saying it's bc of the power outlets near my table, and that is somehow affecting this one, right?

[glacial spruce]

I'm saying that since the circuitry is not shielded, it's picking up interference from the mains wiring in the area.

Since you're working with low level signals, I would recommend shielding everything.

[harsh gust]

ohh how do I do so?

[glacial spruce]

Place all the circuitry in a grounded metal box, basically

[harsh gust]

a-ha! thank you a lot really. Do you see any problem with soldering by any chance?

[glacial spruce]

The soldering looks fine to me, although one of the wires to the jack looks unsoldered.

[harsh gust]

ohh ok I will look into that

[harsh gust]

thank you so much really!

[harsh gust]

but this guy seems like he did it without anh shielding in the end of this vid

https://youtu.be/aOJuCYgmPPE

[vast jasper]

I have a problem with my I2S MAX98357A breakout board from Sparkfun, It works but there weird parasite noise, How can I reduce or event surpress them ?

[fickle leaf]

@harsh gust It’s not entirely clear in the third photo, but it looks like the blue wire isn’t going to the “sleeve” connection ear on the 1/4-inch jack. If that’s the case, all you would hear is 60Hz hum and noise.

Can you post another photo of the jack from a different angle?

[lethal slate]

The docs say you can leave the GAIN pin floating, but that can be a problem. What kind of noise are you hearing?

[harsh gust]

ohhh sorry I checked this message now

right on! I will take a pic right now

[fickle leaf]

@harsh gust I suspect that the Blue wire is going to the Switch terminal rather than the Sleeve. It appears that the Red wire is connected to the correct terminal (Tip).

[harsh gust]

ohh so the blue wire(negative ground stuff) should be connected on the other side?

[lethal slate]

Which board are you using to drive the MAX98357A? It this an MP3 file or WAV?

[vast jasper]

A feather RP2040 and it's a wav file

[lethal slate]

Are you using CircuitPython? Noisy audio has been problem with the RP2040, but I haven't tried a recent build.

[weary gyro]

Is the WAV file saved as 16-bit unsigned PCM? And preferably mono, to start?

[vast jasper]

No micropython, an exemple from https://github.com/miketeachman/micropython-i2s-examples/blob/master/examples/play_wav_from_sdcard_blocking.py

[vast jasper]

PCM_S16LE according to the online mp3 to wav converter I used

[weary gyro]

That looks correct. What is your wiring to your I2S board? Which Feather RP2040 pins are you using? Also, what is your power to the I2S board?

[vast jasper]

11 for BCLK, 12 for LRCLK, and 13 for DIN and for power I use the board 3.3v power but I also tried using my adafruit buck converter with the USB pin and even the 5v from the USB pin

Can I add some capacitor and ferrite rings to meke the sound better ?

[weary gyro]

I don't think that would help. I don't know what the original audio is supposed to sound like, but it sounds a little like your speaker is blown or the power supply cannot put out enough power. (it's distorting mostly on the bass it seems? bass draws a lot more power than treble) Or perhaps the gain settings are too high. Do you have a WAV of silence you can play? If it plays back cleanly, then the problem is not noise that you could filter out, but probably an issue with power

[vast jasper]

maybe it's a speaker problem because it's a big 1.5W old computer speaker and it's a little bit shattered but i had a similar problem trying with headphone jack

[weary gyro]

I believe the MAX98357A is made for speakers and will distort headphones or line-in, unless the gain is turned very low

[vast jasper]

it's the original audio

The speaker is like this

[vast jasper]

i don't have one but I will try

[harsh gust]

Hey I really appreciate this. I just switched the blue wire like you said and it now works properly!!

[fickle leaf]

Super! Let us know how well the stridulation detector works. Would enjoy hearing some footsteps.

[harsh gust]

thank you!

[harsh gust]

May I ask one more question real quick?

[glacial spruce]

You just did 😆

[harsh gust]

lmaooo you got me

So I made this one on breadboard, and perfboard each.

It almost has zero noise. I cannot hear any. But the volume was kinda smaller than I need.

Anyways, it was really strange that even though they are both exactly the same circuit using same components,

the one built on breadboard has smaller volume than the one built on perfboard.

I checked over and over but it was clear that perfboard one had clearly louder volume than the breadboard one.

Why does this happen? anybody knows?

[glacial spruce]

Could be the transistors are different, or capacitance or leakage is absorbing the signal.

[harsh gust]

no I used exactly the same components

[harsh gust]

ohh so is there a higher possibility of leakage when it's built in breadboard?

[fickle leaf]

Solderless breadboards occasionally have contact quality and reliability issues. There’s some built-in capacitance between contact strips, but that usually isn’t an issue at audio frequencies as in this case.

[glacial spruce]

Semiconductors vary a lot, even for the same part number. For an MPF-102, the forward transfer admittance (amplification, basically) can be anywhere from 1600 to 7500 µmhos.

[harsh gust]

ohh interesting

thank you all for great insights, I'm learning a lot from yall

[lethal slate]

Dealing with that variance is an important part of active circuit design.

[median hemlock]

Has anyone gotten audio recording and/or processing working on the CPX or CPB? I’m only coming up with VU meter type projects, which seems like wasted potential given microphone, memory, and speaker are all there nice and tidy.

[odd oasis]

The CPX runs an M0, so it’s pretty limited on memory for recording, and on the slower side for processing audio. I’d imagine the bluefruit has a much better potential in that regard, but haven’t seen an example myself.

[median hemlock]

Those are really good points. Has anyone tested the Teensy audio library port running on the Bluefruit?

[weary gyro]

Teensy Audio Library has only been ported to the SAMD51 as far as I know

[tall mango]

What are some possible fixes for a CD player that recognizes discs, but can't read the tracks or data? I've tried carefully cleaning the lens with a damp soft cloth, and adjusting any potentiometers. This CD player is from the late 80's so I'm not even sure if it's possible to fully revive it. It was working fine yesterday up until one point while playing a CD it just skipped and cut out.

It's a Technics SL-P115

[glacial spruce]

Do you have access to an oscilloscope?

[tall mango]

Not at the moment, but i might know a place that has one if it's necessary.

[glacial spruce]

In that case, I'd probably shotgun replace all the low-voltage electrolytic capacitors (especially in the power supply). They could be dried out and not working well enough. I've fixed a lot of electronics of that vintage the same way.

[tall mango]

Even if they're not bulging or leaking?

[glacial spruce]

A lot of them fail silently. You can test them if you have an ESR meter, but I don't, so I just replace 'em wholesale.

Unlike this mica capacitor from 1928, which is still fine.

[lethal slate]

Whoah check out the magic eye.

[tall mango]

I'm not able to right now but I'll see what capacitors i have when i get back

Hopefully most of them are the same

[glacial spruce]

Yeah, that capacitor tester is a bit of a relic, but it's dandy for measuring things like the leakage of high voltage capacitors, as well as reforming them. It can do other tricks as well, such as characterizing the winding ratios of unknown transformers.

[tall mango]

It would be cool to get a whole setup of vintage electronic testing tools, but if i did they'd probably be outdated or barely working.

My cassette deck barely works as it is because it's so old and i accidentally broke it once only to somehow fix it

[glacial spruce]

During the lockdown, I amused myself by repairing my pile of broken old oscilloscopes.

Now I have a pile of working, but hopelessly obsolete scopes. I'm trying to figure out what to do with them.

[tall mango]

Ooh, i love those round display ones

[glacial spruce]

I had this one display its own name. Not bad for a 75 year old instrument.

[tall mango]

You could make it into a sound visualizer, I've always wanted to do that

Unfortunately i have zero experience with anything that has to do with oscillation, except for sound

[glacial spruce]

I am thinking of making them into steampunk art projects

[harsh gust]

Hello guys! Been awhile.

Today I borrowed the ant colony from a friend of mine, and briefly tested out a simple piezo-preamp circuit on them that I built earlier.

I would like to ask a few questions regarding few issues I faced while I'm testing.

This is the first one. I thought it would be nice if I can get a little bit more gain. There was merely no noise, I could get the sound of ants touching and messing around with the piezo mic. No issues so far.

I disconnected the audio jack for a bit, and put it back on, tried recording again. This time, I could hear this humming noise.

Whenever I touched my macbook, it amplified like this.

This is the crazy part so I suggest you to maybe turn the volume down when you play it.

I tested over and over again and at some point, it started to make crazy noise like this.

The piezo mic was still working, getting the input but there was this crazy noise going on. Anybody guess what is this about?

I'm uploading few photos and video here also.

I took the piezo mic out of nest once it generates that crazy sounds but that is how it looks for now.

This was the piezo-preamp circuit that I used to test it out today.

But I am thinking of making this one for higher gain boost and noise control and test it out.

Could anybody help me out reading this circuit diagram?

I already prepared all the components I need for this one but it takes so much time for me to read it off of that kinda diagram to make one.

If anyone can make easier version of that with sth like Fritzing, I'd really appreciate it.

[glacial spruce]

Probably one of your connections came loose

Or you were touching something with an electric field. Macbook power supplies come with a 2-prong plug and a 3-prong plug. You may want to try the other kind to reduce hum.

[fickle leaf]

Sometimes recording with the laptop running on battery only can help to reduce audio recording noise, too.

[harsh gust]

what would be the best way to check which part of the connection causes problem? and what is the best way to prevent it?

[harsh gust]

touching sth with an electric field? could you elaborate a little more? you talked about shielding cables and the components and stuff before, right? would shielding them and encasing them solve the problem?

[harsh gust]

how is this noise issue related to whether using 2-prong plug or 3-prong plug?

@glacial spruce @fickle leaf But as yall can see in the photo, I wasn't even using a charger while I was recording.

[fickle leaf]

Are you feeding the signal into a 3-pin input connector (XLR)?

If so, that could be introducing some noise since one of the 3 pins on the 1/4” connector would be left open.

Is a 1/4-inch “instrument” input connector available?

[harsh gust]

oh! I'm using 1/4" to XLR(3-prong)

so the 1/4" side is connected to the circuit and XLR is connected to my audio interface (apogee duet)

yeah 1/4" input connecter is available. I just used that cable just because that was the first cable I could find around me 😂

[fickle leaf]

That’s called a TRS to XLR adapter. Won’t work well in this situation. A regular 1/4” to 1/4” should work better.

[harsh gust]

So using XLR(3prong) socket on one end might be causing the humming issue?

[fickle leaf]

Yes. The XLR input pins are +signal, -signal, and ground. One of the signal inputs is left open using that cable.

If you look at the 1/4” side, you’ll notice three connections, the Tip, a Ring, and the ground Sleeve. TRS

It’s currently plugged into a jack that only touches the Tip and Sleeve. TS

[harsh gust]

yes it is correct!

oh so it that also causing that crazy distorted noise?

[fickle leaf]

Perhaps. The open connection might be acting as a noise antenna.

[harsh gust]

that's interesting.

[worldly copper]

where does one come across such a pile of broken old oscilloscopes to fix? for a friend...

[glacial spruce]

Hamfests mostly, but I also have the word out that I'm interested in odd old electronics, so people tip me off to junk sales and the like.

[worldly copper]

hamfests? and ive got to make some electronics friend irl so they can tip me off to these things too lol

[glacial spruce]

Normally I'd suggest going to the arrl.org website and looking at their list of hamfests to find some near you, but their site is down for maintenance at the moment.

[solemn flint]

The local electronics swap meet has been defunct for the last two years because of covid restrictions, but it's back next month! 🙌

[azure panther]

Nature...nature is healing.

[fair forge]

Can y'all check me here? I'm doing a continuity test on the mic... I put the black lead on the case of the mic and the red - if it's positive, it should still be 1, correct? Thanks

[glacial spruce]

I don't understand your question. The red lead to what? 1 what?

[fair forge]

Nevermind... I figured it out. Forgot I posted this.

But to satisfy curiosity... I'm using a multimeter to check polarity and I forgot if one was the pos or neg if the black wire on the multimeter is on the case of the mic.

[tall mango]

Is there another reason to consider as to why my CD player is failing? I've tried relubricating it, replacing capacitors, cleaning the lens, adjusting potentiometers, and loosening/tightening the laser assembly. It recognizes the disc is there, but it can't read anything off of it. It doesn't have any belts or gears in the actual assembly, it seems to be using magnetic coils to move the laser head, and same with focusing the lens. It's a Technics SL-P115 and has a SOAD32A laser assembly in it, which I can't find online at a cheap price.

[random bone]

are you sure it's spinning up? https://audiokarma.org more than one person has motor troubles. The lasers can go bad too, I've heard of that.

interesting no belts. I have fixed my elderly Sony player by replacing the belt (first time with a matching rubber band).

[tall mango]

The only belt is in the disc tray eject mechanism

I just tore down the laser assembly completely and wiped down the lens components, I'm putting it back together now

[random bone]

i have not seen them get dirty, but it depends on the atmosphere in your house

[random bone]

yes, now that I remember, maybe that was just eject also

[tall mango]

Now it just won't eject

Which is strange, i didn't touch the eject mechanism

Also the laser focus lens isn't moving up and down anymore, it's moving side to side

[whole pasture]

question for y'all

I have two of these

one has the proper adapter for my headphones, and one has a good mic

I'm tempted to just snip off this bit, swap it, and solder+heatshrink

will that work?

[azure panther]

If you can figure out which wires go to which, it should

but those wires are very small.

I've tried to work with them before, I gave up and found another solution.

[whole pasture]

oh I've successfully soldered tiny audio wires before (making use of that youthful dexterity while I still have it lol)

I'm just wondering if there's any weird amplifier stuff that I gotta watch out for

[azure panther]

I don't think so? But my audio experience isn't super extensive.

[whole pasture]

Update: janktastic, but it works!

(buying these two units & frankensteining was STILL cheaper than a good mic that works w/ my headphones)

[azure panther]

That's great!

[golden lark]

Is there anyone that could help me make a spectrum analyzer for some iregularly shaped RGB matrixes? All I know is I need 35 bins and a way to map the loudness from 0 to 25. (specifically looking for a fourier transorm)

I originally started with this https://learn.adafruit.com/fft-fun-with-fourier-transforms/spectrum-analyzer but attempted to draw the bars in columns.

[glacial spruce]

Looks like that link covers the complicated bits (Fourier), you just need a mapping from the output range to your LED column size.

[mental axle]

Heya. Im currently tinkering with a mems microphone and to check the quality i first plugged it into my cp2615 devkit.
the quality sounds a bit robotic. Any clue why? inmp441 is the mic. maybe it's just not that good?
recorded at 44khz 24bit

[dull basalt]

doesn't sound that bad except the mic need a foam cover for sure...

[mental axle]

No it's not bad. But it doesnt sound 16bits 44khz good.

[glacial spruce]

It could be your playback hardware doesn't support that resolution and sampling rate. If it's resampling it at 44.1kHz, there are going to be artifacts.

[mental axle]

im using cp2615 i2s-usb bridge

it can do 16/24 bit and 44/48khz

24bit 44khz should be the correct values for the inmp441. i tried 44 and 48 and it sounds the same :/

[young spire]

I guess I'm just not hearing what part of it is robotic?
It's certainly not a large diaphragm condenser mic in terms of quality, but that's a given considering it's MEMS.

[mental axle]

i heard a comparison recording from this mic, on youtube which does not show the robotic part. so i know it can do it without. just trying to figure out how to get rid of it

[glacial spruce]

You're comparing 44kHz/24 bit audio with heavily compressed youtube audio?

[mental axle]

yesnt. i know that youtube does not compress this artifact away, because i observed this robotic noise i mean also in other youtube videos

i can send you an example if you would like

[young spire]

Probably some filtering they're doing along the way, but the fact that you've heard whatever this "robotic" sound is in another recording from another device, seems to indicate something inherent to either this product or your playback equipment, rather than something malfunctioning with your unit.

[mental axle]

Probably. Or it is the cp2615 that has some known incompatibilities as i now found out.
My playback equipment is verified working perfectly.
I will just get another mems or an adc directly and try that with the cp2615. Thank you :)

ah yeah. i almost forgot. i just tried to get the esp32 hooked up as a noise/sine generator to the cp2615, but i just cant get them to talk to each other. (esp32 as i2s slave)

@young spire

[olive cairn]

I got an audio fx board with 2x2w amps. I wired it up to a single speaker and tested the pre-loaded audio file by shorting trigger pin 0 to ground. It worked fine then. When I soldered my wires to a button, most of the trigger pins appeared to become shorted to ground permanently. The trigger pin caused a constant loop as if the button was held down.

I desoldered the trigger pin 0 and ground wires. An ohmmeter shows between 400-1700 ohm resistance between the ground pin and most of the trigger pins. I can't see any visible short. Any ideas on what I did wrong and if I can undo it?

[glacial spruce]

It could be that a short circuit or static electricity damaged the chip

[olive cairn]

Thank you. Are there any other obvious possibilities? That doesn't seem like an easily fixable problem.

[random bone]

You are measuring internal resistances on the chip. I tested with the same or a similar board, and I see more like 20k. If you reverse the test leads, do you get the same measurement?

I think this is likely to be a wiring problem. Now the that button is removed, does it no longer work at all?

A clear photo of the pins in question and both sides of the board would help.

[olive cairn]

I desoldered everything, and the resistances were around 400 or 1.7k for most/all trigger pins to ground. I'm pretty sure reversing the leads gave the same results, or maybe that's what is causing variation between 400 and 1.7k . I believe the resistance is supposed to be around 100k. I can get a photo when I go home. I wiped it with a lint-free wipe and used copious compressed air to blow off any stray particles that might be shorting connectors, but neither of those did anything.

[glacial spruce]

Is this powered or unpowered?

[olive cairn]

I measured the resistance unpowered. If it's powered, the circuit appears to be closed regardless of any button press or deliberate short by connecting bare wires to the pins.

[glacial spruce]

It could be that what you're really measuring is the conductance of the protection diodes.

[lyric galleon]

Hey guys, right now I'm attempting to hack in an aux cable into the power amplifier section of my old sony icf-9650 AM/FM radio. I have a schematic of it (I can post if it'll help), and have determined the following:

1. There is a white cable going from the main board to the tone-control (bass/treble knobs). This contains the audio signal that is roughly 0.2Vpp (measured with my oscilloscope).
2. There is a 0.022uF capacitor connecting the IFT cans and the ground plane on the tone-control board. I'm assuming this is some sort of noise suppression stuff?
3. I disconnected this white cable between the boards and connected it to the audio signal cable from a mono 3.5mm cable. I plugged this into my raspberry pi. The voltage level coming out of the pi is roughly 0.3Vpp.
4. I connected my audio ground to the general ground of the radio (seems to be shared by everything).
5. I cannot hear the audio coming out of the speaker. If I crank the volume ALLL the way up, I can faintly here the audio.

Any ideas?

[glacial spruce]

Could be a signal bias issue: try connecting it before the coupling capacitor

[lyric galleon]

lemme see if i have another 0.47uF capacitor lying around

because the board is fairly inconvenient to access directly due to all the mechanical capacitor tuning stuff

[glacial spruce]

It doesn't have to be exactly 0.47µF. A smaller one will work, but you'll have less bass response, a larger one (within reason) is fine

[lyric galleon]

i've got a 0.1uF cap, should work for just a test

[glacial spruce]

Might be a little tinny sounding, but should let you know if it's likely to help.

[lyric galleon]

THAT DID IT

YOU SAVED MY PROJECT

THANK YOU

would it be a sin to just toss 5 0.1uF decoupling caps in parallel and call it a day

or is there anything more special about "audio" caps

or is that magic fairy dust

[glacial spruce]

No, that should be completely fine, as long as they're not high-K ceramic.

[lyric galleon]

they're high-k 😬

i'll just order a film cap kit off amazon, arrives tomorrow 😄

[olive cairn]

I took another crack at it. I threw a 10k resistor on the ground side of the switch for trigger pin 0. The board now only thinks the 0 pin is triggered when the button is pressed, so that's an improvement. On the other hand, it appears to not work for the hold-down-to-loop functionality ("TnnHOLDL.OGG" files). But I hadn't fried the chip, so that's good.

[turbid shoal]

I'm looking to fix 2 SMD LEDs on a beat step pro midi controller. Can anyone tell me best way to figure out the LEDs type or is there a schematic?

[glacial spruce]

I have no idea on a schematic, but there aren't that many kinds of LEDs out there. Do you know what color they are/were?

[dull basalt]

got a pic / model+make of the midi controller ?

[keen cradle]

HI Guys! I have been trying to get this display to work for the last 2 days and i am my wits end.

ITSBitsy 52840 express W Adafruit 2.13 Featherwing 1680 eink display

I load the code, Pins match their respective numbers in the code. But when I run it I get nothing.

I linked the wiring photo below

I want to double check to make sure I am fully wired up properly, what can i do for testing where the problem is?

[random bone]

Since it's a FeatherWing, if you have a Feather board of some kind that can run this code, you could just plug that in directly and see if it works

[tall mango]

Anyone know of a good compact USB audio DAC IC? something around 48khz stereo should work

[median obsidian]

So I just made my own stylophone but there seems to be something wrong with it

New to electronics, so I dont know where to start for a diagnosis

If anyone can help out that would be great

It was working fine for a bit then started making these tick noises.
Since the speaker is making noise I assume the speaker is fine and the connection are fine. Maybe a cap or an IC shorted or something.
No clue how to go about finish the problem. Any direction would be great

[zinc palm]

I have a need for audio beat detection with CP and am wondering what the best approach would be. I'm using the Arduino Nano RP2040 Connect which has a PDM microphone, so I can sample audio data on a time basis. My thought is to pass that to a FFT processor to get a frequency breakdown and then analyze the resultant data interactively to get the necessary info. I've seen some options for doing this with Arduino but haven't come across any CP implementations yet. Anyone have any suggestions? Thanks.

[glacial spruce]

To get the beat, you may be able to use a simpler averaging/envelope detection algorithm instead of something as complex as FFT.

[zinc palm]

K. Can you point me in a direction to learn more about using the averaging/envelope detection approach?

[glacial spruce]

There's a slide deck that explains the concepts here: https://resources.mpi-inf.mpg.de/departments/d4/teaching/ss2010/mp_mm/2010_MuellerGrosche_Lecture_MusicProcessing_BeatTracking_handout.pdf

This thread includes a link to possibly useful source code: https://www.avrfreaks.net/forum/music-beat-detection-lilypad-wearable-target

[lucid oracle]

I'm having quite a bit of trouble getting this microphone to work at 16 kHz sampling rate. I can only get it working up to 12 kHz. Has anyone else had success with this?
https://learn.adafruit.com/adafruit-i2s-mems-microphone-breakout?view=all

[odd oasis]

When you say "trouble getting this microphone to work," what kind of errors or bad data are you getting?

[lucid oracle]

At 14 kHz sampling rate, the microphone outputs normally for a few seconds (maybe up to a minute) and then I get errors in the console (not sure what it's called, but the area at the bottom of the Arduino IDE that prints the uploading messages, etc.) to the likes of "java.lang out of heap." or something to that effect. This is making me think it's some sort of memory problem, but I'm not sure why that would be happening. I'm not storing any data in my code, just printing it directly to the serial monitor. I'm thinking maybe it has something to do with how the I2S library is handling things? Maybe I'm not clearing the buffer as fast as data is coming in. I would imagine that would be handled by the library.

Any sampling rate larger than 14 kHz and the data prints for a few less seconds than it does at 14 kHz (maybe for 10 seconds or so) and then stops printing altogether. No errors show up in the console area (not sure what it's called, but the area at the bottom of the Arduino IDE that prints the uploading messages, etc.)

I'm using a SAMD21 MCU. SparkFun's mini breakout version https://www.sparkfun.com/products/13664, but Adafruit's microphone breakout.

[smoky cairn]

Hi all. Any Linux/ALSA audio experts here? Asked on #help-with-linux-sbcs #help-with-linux-sbcs message

[tall mango]

I have a bit of a problem, i got a Toyota stereo to work as a home stereo system, and i recently rewired it with a removable connector. The problem is now i don't hear a thing! I had L+/- and R+/- wired to the RCA connectors accordingly along with SGND on both ground terminals and i heard nothing. I tried disconnecting the L and R minus terminals but no difference

[glacial spruce]

L+ is one pin, L- is another pin, and SGND is a third pin. RCA connectors only have a pin and a shell. How are you wiring these again?

[tall mango]

Right now L+ on the left channel, R+ on the right channel, and SGND on both channel grounds

[glacial spruce]

I wonder if the left signal is between L- and L+ instead of between SGND and L+

[young spire]

found a post online indicating SGND is for Fade? But like madbodger, curious the schematic of your prior wiring that worked

http://www.mr2.com/forums/threads/15667-Toyota-OEM-Stereo-Wiring-Information/page2

[median prawn]

I want to have a highpass/lowpass fade between two signals controlled by a knob, what's the best way to do this?

just use a voltage controlled resistor in a RC filter? I ask because I've never used a VCR before

I guess I really want to know how to use a VCR, on Wikipedia there's only one schematic and it looks like I just need an input voltage and then I get a resistance between two pins, is that all there is to it?

[last prawn]

In terms of the audio, what is the goal? aka: When the knob is at 50% what do you expect? When its at 25%, 75%?

[median prawn]

It to be at a given frequency, I can calculate what frequency it will be at on my own, that I know how to do

[last prawn]

ok, but like ,at 50% do you want both signals fully on? or is it ok that signal a is always lowpass and signal b is always highpass?

[median prawn]

both filters should be the same frequency

[median prawn]

Always lowpass/highpass

[last prawn]

Cool, what you want to look up is a Crossover 🙂

(There is a DJ effect that is similar to what you're describing but a bit more complex)

The DJ effect isn't a crossover

[median prawn]

Cool I'll take a look at the circuits I find for that, I probably should've searched that but I wasn't sure enough on the term to put in the effort

[last prawn]

https://en.wikipedia.org/wiki/Audio_crossover

[median prawn]

Thanks!

[glacial spruce]

Also "tone control".

[median prawn]

👍

[fair forge]

Hi y'all!

I'm running the following code on an ESP32-S3. The code works fine, but doesn't quite do what I want it to do. What I want is for all the LEDs to turn on simultaneously with sound activation... but this code moves more like a VU meter. Also, I want all the neopixels to be white, not yellow/green. I've been changing variables for a couple days now but can't seem to get it to do what I want. Can anyone help? Thanks!

from rainbowio import colorwheel
import board
import neopixel
from analogio import AnalogIn

led_pin = board.D6  # NeoPixel LED strand is connected to GPIO #0 / D0
n_pixels = 32  # Number of pixels you are using
dc_offset = 0  # DC offset in mic signal - if unusure, leave 0
noise = 100  # Noise/hum/interference in mic signal
samples = 60  # Length of buffer for dynamic level adjustment
top = n_pixels + 1  # Allow dot to go slightly off scale

peak = 0  # Used for falling dot
dot_count = 0  # Frame counter for delaying dot-falling speed
vol_count = 0  # Frame counter for storing past volume data

lvl = 0  # Current "dampened" audio level
min_level_avg = 0  # For dynamic adjustment of graph low & high
max_level_avg = 512```

Full code...

[glacial spruce]

To get white, you probably don't want to use colorWheel(), just set the pixels to white (I think that would be [255, 255, 255])

[fair forge]

Thanks @glacial spruce ! I got that one, but now have no idea how to make them all react simultaneously.

[glacial spruce]

I'm guessing there's a call (or array slice notation) to "set a bunch at once" or alternatively set them then do a global update. However I don't know what those methods are. It might be worth asking that question in #help-with-circuitpython

[random bone]

I'm not sure exactly what you want to do, but when you create the NeoPixel object, you can set auto_write to True or False. Default is True. If False, the strip will not be updated until you call .show(). So you can make all your changes and then do .show(), and the changes will appear to happen all at once.

If you want to set all the pixels to a single value, use .fill().

[fair forge]

I'll give that a shot when I get home... Definitely a path toward my understanding of this. Thanks!

[tall mango]

For splitting RCA signals, do i need an audio distribution block? They all seem really expensive for what I'm assuming is just some RCA connectors soldered in parallel.

[glacial spruce]

You don't need that, just some Y cables (or just cut some up and attach them yourself)

[tall mango]

It gets kinda messy, i can't just keep stringing Y cables together for all of my devices. Which is why i was wondering if there's some kind of splitter block that isn't super expensive.

[glacial spruce]

You could, of course, do just what you mentioned and solder a bunch of RCA connectors in parallel. Arrays of jacks are available inexpensively at many surplus vendors, such as this one: https://www.goldmine-elec-products.com/prodinfo.asp?number=G23612

[tall mango]

Ok I'm gonna try that. Hopefully I can find enough RCA jacks.

[glacial spruce]

If you have any multichannel audio amplifiers in your junk pile, they tend to have nice arrays of RCA jacks.

[tall mango]

I do have an old VCR that I basically killed for fun when i was younger

Poor VCR

[young spire]

Depends on what you need for signal integrity.

You don't need to worry much about reflections like with video signals, but if you have a 16-way passive split, you're likely to notice some signal degradation, especially if you're powering speakers directly from that RCA

[rare fog]

Hi

I'm trying t play wav files on a Rp Pico using an Adafruit Max amplifier

it's working ok, but I get loud static sometimes

I'm thinking it might be caused by uneven power supply

Is there an easy way to introduce some capacitor or something into the board to even out fluctuations?

[bronze locust]

should I expect the sound quality from a $10 USB audio adapter (which as I understand it is a DAC) to be worse than the onboard audio of my ThinkPad?

[weak zealot]

complete tossup tbh
as a broad note apple's usb c to 3.5 jacks are actually really high quality DACs and they're like 12$

they beat a handful of several hundred dollar units

[bronze locust]

a friend just told me that. What is going on there... why are they so much better than the rest of the industry?

is apple operating at a loss to attract customers to its white picket fence product line?

[weak zealot]

It's pretty interesting!

https://www.audiosciencereview.com/forum/index.php?threads/review-apple-vs-google-usb-c-headphone-adapters.5541/

[rare fog]

I solved my static problems with the Rpi Pico by feeding the Pico and the Audio board directly from the power source. In the examples from Adafruit and others they feed the Audio board from the 3.3v out on the Pico, but that caused lots of static while the Pico was working

Strange that they wire it like that in the examples. But maybe it's my Pico that is faulty somehow

[scenic axle]

Yeah so apparently the one bad thing about the apple dongles is that the amp isn't super powerful, but as long as you're not trying to drive something super crazy with them it'll work very very well

[bronze locust]

Just picked up this baby open box for $5.60 and plugged it into my powered stereo through a USB A to C adapter. Mwahaha, I have higher quality audio than my ears know how to tell the difference between!

[scenic axle]

Awesome! Yeah I used to be really big into the audiophile scene and I remember the apple dongles regularly sounding better than lots of DACs 10-20 times their price

[bronze locust]

Yeah I'm still truly baffled at that

[scenic axle]

It happens every once in a while. I'm lucky enough to own two pairs of genuine Sony MH755 IEMs (fancy word for earbuds), which were sold for $7 but regularly outperform IEMs that cost well over $1000

I own an ungodly number of headphones and earbuds and they are some of my favorites despite being the cheapest I own by far

[bronze locust]

what's the thing that happens every once in a while? that a high quality product is cheap to make, or that a mfr underprices a product?

like why is it that apple is able to sell a high quality device much cheaper than any other mfr does?

is it that they already have the moat of a high quality manufacturing process so it's cheap for them to produce products that are expensive for smaller companies to produce? and if so, shouldn't their bigger competitors also be able to do comparably?

[scenic axle]

Well I think Apple was able to do it because they have the capability to manufacture things very cheaply at large scales and have engineers that really know what they're doing

The sony thing was much more of an accident AFAIK

[cobalt rain]

Sony is the Apple of Japan. Their good stuff is really good.
My MDR V600 headphones are my reference standard.

[scenic axle]

Oh yeah, their stuff is amazing, I just feel like making $7 pair of earbuds perfectly tuned to the harman target was unlikely to be entirely intentional

[mint laurel]

I'm trying to get my RPi pico to play a wav file. I'm using an audio expansion card (https://www.waveshare.com/wiki/Pico-Audio) and I can't seem to get sound to come from the speakers. I've tested them though, and they work. Here's the code I'm attempting to use:

`import audiocore
import board
import audiobusio

wave_file = open("BingBong.wav", "rb")
wave = audiocore.WaveFile(wave_file)

audio = audiobusio.I2SOut(board.GP27, board.GP28, board.GP26)

while True:
audio.play(wave)

while audio.playing:
pass
`

Is this the right place to be asking for help with this?

[lucid anchor]

Welcome! You'll probably get better help in #help-with-circuitpython, as it appears that's the language you're using. We discourage cross-posting in genera, but in this case, it's alright to post in the suggested channel. Good luck!

[mint laurel]

I appreciate it!

[lethal slate]

Take out the "while True: " line. You're restarting the clip before it has a chance to produce any sound.

[last prawn]

will it finish playing if the loop terminates? I have a feeling the 2nd while loop is supposed to be indented

(assuming this is supposed to play the sample over and over)

[lethal slate]

Yeah, I saw that this was discussed in the other channel.

[solar jasper]

Any good, easy way of adding rca audio to a pi0?

I know how to add video. is a hat going to be the easiest and most space saving?

[last prawn]

I think there are hats with RCA built in - you might save more space with a different kind of output and an adapter though.

[young spire]

Easiest is probably just going to be a cable adapter. Depending if you care about vertical space or horizontal space will determine that vs a HAT

[lethal slate]

https://learn.adafruit.com/introducing-the-raspberry-pi-zero/audio-outputs

[azure panther]

This is annoying. I want to read the article but any link I click on adafruit's site leads me to the login page

[glacial spruce]

Yeah, I get that sometimes. Then I gotta go find my phone.

[weary gyro]

I have that problem too. Some weird stale cookie. Clearing the cookies for *.adafruit.com let's me view pages without logging in

[terse maple]

i wonder if the recommended non-DAC audio output should be some kind of delta-sigma modulation or PDM? PWM audio seems to need a lot of filtering. though you might need a MCU with a peripheral I/O state machine like PIO to do it with an acceptable CPU load?