If anyone is contemplating using the D1 mini form of the ESP32, you'll need to change some of the pin assignments in audio_reactive.h before compiling. I used the following (also mentioned on the webpage for the WLED fork): #define I2S_WS 26 // aka LRCL #define I2S_SD 18 // aka DOUT #define I2S_SCK 22 // aka BCLK Scott: trivial point but the web links for a couple of the required libraries at the top of your sketch are swapped around.
Good to see a man after my own heart - using spare network cable as a plentiful supply of wire. Neat and tidy project. Pro-tip - broken flat screen TVs are a good supply of diffuser sheets.
Brilliant work. I got a bit silly and made one 48 wide by 16 high. All because as a kid there was always a red led spectrum analyser in a shop window on Oxford street. I was fascinated by it and how it worked. Always wanted to build one and this project gave me the inspiration. I bet that one on Oxford street cost an absolute fortune back in the late 80s.
Very slick Scott - Fast paced with lots of useful content. Battery power and mic with I2S interface. Also loved the way you didn't stand on ceremony with the soldering - no PCB just flip the IC and solder the wires straight to the ESP32 pins. Fit for purpose and gets the job done👍
Thanks for your comments :) for something like this a PCB would be more trouble than it's worth. It would be nice to have a custom PCB though that integrated everything with headers for the mic, and a charge and boost circuit all on board. But I'm not that good at electronics I'm afraid!
Also, I loved the way you side stepped the captive portal. You're solution for your purposes seems much better especially if other members of the family want to use it hassle free. Personally, I feel like the captive portal is itself a fiddly workaround. Why not use a cam enabled ESP32 and get it to read your network QR code.😊
Hey, someone has just made what I suggested www.reddit.com/r/esp32/comments/lvlhl1/qr_code_for_esp32_web_server/? OK - how about adding a time machine to it!?
This looks fantastic! It's such a beautiful, clean looking display and your patterns look so nice! I love how you display the IP Address on startup. That's a really smart decision. Thanks for the creating this vid with such clear and easy instructions, and also for the link to your code.
as usual, a great video ! I discovered your channel a few weeks ago and only 1h after finishing the first video I ordered my first esp and led strip ! I now have a working prototype and i'm working on a box which can transmit signal to the VU display over wifi so that your sound source does not have to be close to the display (I'm not using a microphone so I only display the music and not the surrounding noise). Thanks for the work you've done, can't wait for the next project !
Sounds interesting! Are you sending the actual audio data or doing it how WLED works and transmitting the results of the FFT? If the first way, is there any latency?
having just gone through the setup of my first couple of esp32's and doing a bit of playing, your videos have been such a help... to the extent my kids even said I was cool tonight. Im not telling them it was you but kudos and keep up with the vidoes. Thanks
That's very kind of you! I've got a bunch of things I'd like to do at the moment but lockdown in the UK along with too much 'real' work is making if difficult. Ah well, will get there eventually.
Thank you very much for your videos, I have collected your projects. Everything works very well. I modified the sketch a little and got very nice effects.
This one runs much faster than the other one you commented on as its using the second core to do all of the audio processing. It's a good idea to do this I think!
4:55 you mention you had to remove the usb socket, but didn't went into detail as to why. Great content as always Scott. Looking forward to your next project.
Nice one! Seems like a solid design. I think I'll build it. I do a lot of pixel art, having a little frame like this running a server I can drop gifs onto sounds like a very fun project.
To help make sure you don't run out of ideas :) How about a "nano leaf" project? For a little twist on the theme maybe some sound reactivity for activity around it.. ie some sound of a person in the room.. mellow animations.. some music maybe slightly more reactive and heavy metal.. very reactive.
It would be a very good topic for youre fastled Series where You explain how to use multiple groups and eventuely with different Animations or slitly different 👍🏻😍🔥
Hi Gabriel, what do you mean by groups? Are you talking about multiple strips connected to different controller pins, or groups of LEDs on the same strip or perhaps both? Either way, it's a good topic for a future video, and if I can find the time I might well do it!
@@ScottMarley Yeah, I mean both variations. What I am interested About is how You can animate lets say 5 groups in one strip simultane without or minimum delay. The Same About multiple Strips At different Pins. Thank You Scott!
Thanks for all the work you put into this awesome project. I put the 16x16 version together over the past few days and it came out great! I had many of the items on hand and only had to buy the display and microphone. Including the Fusion file was especially helpful as the acrylic sheet I had was not 2 mm thick. (Note that if you change the "Glass Thickness" parameter, you need to adjust the "Grid Thickness" accordingly). I did initially try 3mm translucent acrylic, but the display was blurry looking. I switched over to 2.5 mm clear with the printer paper behind it. The paper produces a little pulp pattern but the projected image is much sharper. Ultimately, I may try to find some thin translucent film to put behind the acrylic.
Great, so glad somebody else made one! You're right, there is a fault in the file as the glass and grid thickness should reference each other. I didn't notice this as I never changed that parameter to find out. Glad you managed to figure it out anyway!
@@ScottMarley That seems to be more of an issue than I originally thought. I looked at a number of other papers that either have watermarks or the same fiber-ey pattern. Any white vinyl sheets I have found online are adhesive backed. At the moment, I'm using a piece of white UltraCote model plane covering material in the place of the paper. It's far from ideal. It likes to roll up, so it's really fiddly to get into place. It also has a heat-sensitive adhesive backing, so I'm not sure how it will work out in the long run.
Thank You for your work on projects, and sharing! One question / suggestion: would be very very nice to still have some physical button(s) controls option. I also put this suggestion on Github. I will try to add that myself, but, not sure will I succeed :) I am more hardware guy. Thanks!
I love this! And your videos are great, thank you Scott! I'm having some trouble though I hope you can help with, I've connected everything up and everything seems fine, except there's just no reaction to any sound. If I touch the SD pin on the mic, the UV jumps on the matrix, so it seems they are able to communicate. I can't for the life of me work out how to downgrade to ESP32 core 1.0.6 and I can't find any info online - could you point me in the right direction? Or do you have any other suggestions? Thanks!
Scott, first I would like to congratulate you, you do amazing things. I speak here from Brazil and I follow all your videos. I built a matrix that worked perfect, but I can't get the same pattern from your video. In this video pattern, the leds start to light up at the base and have this drop effect. Is this pattern in your program available on github?
I'm not completely sure which pattern you mean, is it the one that's mainly blue? That uses the createWaterfall and moveWaterfall functions. Basically you color the bottom row depending on the intensity in each band (red highest intensity, blue lowest, createWaterfall) then every so often just move each row up (moveWaterfall). If you want to integrate it into your own code, you'll have to understand how mine works first, or it might be easier just to write your own version of it!
Awesome❤️❤️❤️.....also bro I said I ll try to make attachable PCBs for the led cube.....I could come up with the idea but making the pcbs Is beyond my level of expertise.
I've watched a lot of your videos and they are all brilliantly detailed. I'm building this project as my very first go at electronics, so I really appreciate how much care you have taken with this work. So (knowing nothing!) ... the code comment says 'LED data, D2 via a 470R resistor', but your wiring diagram doesn't show this. I'll follow the diagram :) Thanks Scott.
I've done that on previous builds and it works fine and lasts forever! I wanted to investigate internal power this time. It sort of works, but I learned a lot doing it so that's more important!
@@ScottMarley yes thats true... I did also Use tp4056 charging Modules whith Protektion and an external Boostconverter for my Projekts. But i like to use cheap Powerbanks whithout Batterie, to recyce old li-ion cells from dead Laptop packs. Mostly only 1 or 2 out of 6 cells are dead. For a 2s charging board i use the "X-lite charging board" (its for a remote) it charges and Balancing the cells from 5v. its a little extensive, But its the only Module i know that Balance Charge a 2s pack from 5v.
Hopefully one day. The electronics parts of TH-cam is fairly crowded, there are a lot of people with amazing content that I have to compete with but I'm trying!
Thanks for your comments! I'm not that good with Fusion at the moment so I don't feel too comfortable with making tutorials yet. There are plenty of really great channels out there that already do it really well though.
Awesome work, thanks for sharing A question, but don't thing it's a small question lol.. Do you have a video on how to wire and program an SD card to ESP32, for example, to play music and lights effect files and pair it to multiple ESP boards through wifi, thanks
I'm afraid I don't. If it's just syncing light patterns up, you can use WLED which can sync multiple esps over wifi. Then perhaps have another on in charge of the music? You can get SD card readers with a built in audio output that talk serial to any microcontroller (I'm making one at the moment that runs from a nano). Maybe look into those devices?
Hi Scott, thanks for more amazing work. I think this is going to be my next project, but will likely use WS2812b strips (not the matrix). I would like 16 pixels wide but 24 pixels tall. Can you confirm that modifying M_HEIGHT to "24" will accomplish my objective? Thanks again!
@@ScottMarley Hmmm. The matrix I built is 16 columns wide, each 20 LEDs high. I change M_HEIGHT to "20". For unknown reasons, the VU meter behaves like there are only 16 LEDs in height. Very strange. Still investigating how to convince that there are actually 20 LEDs in each column. edit: Well, it turns out that I made a bad assumption. I am building this with strips of WS2182b lights, not a pre-built matrix. I assumed that the matrix was wired in a zigzag pattern starting from bottom left, going up, zigzagging back down, then repeating. I assumed wrong. The zigzag is horizontal. In the sketch, I changed "HORIZONTAL_ZIGZAG_MATRIX" to "VERTICAL_ZIGZAG_MATRIX". Issue is resolved. I also had to add 4 more zeros to uint8_t peak[ ], uint8_t prevFFTValue[ ] and uint8_t barHeights[ ] Finally, one last thing: If you want to start your Data in (DIN) from the opposite side, and you explain the code, you can adjust M_WIDTH to a negative value, but please be mindful that it causes an issue with the formula for NUM_LEDS. That number cannot be a negative value, so replace the formula with the exact number of LED (corresponding to M_WIDTH * M_HEIGHT). Cheers!
Good day, thank you for the video. I repeated This project works very well, but I want to ask Cannot connect the button. If not, please add this feature, thanks a lot.
Hi Scott, while i verify the code it shows me an error, please help me fix out. Compilation error: 'cLEDMatrix' does not name a type; did you mean 'LEDMatrix'?
Scott, I want to make a spinning persistence of vision (pov) display with fft. Have you thought about something like that? My vision for it would have an oscilloscope like output of the audio waveform in at least 3 and likely more bands. It would show only the peeks rather than a vu like output so you could see the trace of the waveform for each band. It would probably use bluetooth or wifi whichever I could get the lowest latency on. I'd want the whole controler plus LED strip to be spinning and power to get picked up through something like an inductive pickup so it could run all the time instead of having to get charged in a couple hours. I think this matrix with fft is very close to at least the LED driving that I want to do. I'd like to have your mind thinking about this. What do you think? Oh one last thought is that all the bands would have their moving peek of a different color sharing the strip so they'd be superimposed. The colors would have to combine if the peek appeared on the same LED. I was thinking say, 64 LEDs in the strip, or whatever the densest number one can fit in say 1/2 meter. Mike
Good idea, are you thinking something like this create.arduino.cc/projecthub/theSTEMpedia/persistence-of-vision-pov-display-using-arduino-583d5f but with addressable LEDs? If the power is inductive, it would have to be a very powerful transmitter. For 64 LEDs you'd want at least 2A - even the best wireless phone chargers only just get near that and they're not moving. You lose a huge amount of efficiency if the coils aren't exactly lined up. I think you could get away with far fewer LEDs though, so might be doable with a wireless phone charger type arrangement. The patterns would be relatively easy compared to solving this issue I would think. I wonder how WiFi or Bluetooth would work when spinning round that fast?! I think it would be fine but would need investigating. The audio sync stuff over WiFi is already integrated with WLED, you could have a box somewhere with an ESP32 and a microphone sending the data over WiFi, and another ESP32 listening to the data and running the display. Just my initial thoughts, I do like the idea though and I might keep it in my head until I have some time for another project!
@@ScottMarley ok i do like the esp32 sending data over wifi idea, i also thought about slip ring power but didnt like the idea much, maybe i just need a big big battery? I have a windows media player visualization that does the fft and puts a scope in up to 6 bands. I've noticed even withing windows any lag really puts you off visually connecting the audio. So yeah I'd really worry about latency.
@@jmcarp0 I do think all this needs to be investigated further. You just made me order some wireless charging coils from AliExpress to experiment with ;) I won't have anything any time soon though, really busy with real life unfortunately.
Well I did order the esp32 2 pack and the microphone, and a 1 meter (144 LED) strip of ws2812b. I feel pretty confident I can get that going whether or not its spinning with big battery.@@ScottMarley
Hey! Thank you for this project! I have a question. In my Arduino IDE, there is one error that keeps popping out: "Error compiling for ESP32 DEV Module", do you know how to solve it? It would help a lot!
Scott! Thank you for all the info! Really great project! I am trying to do a similar project with a big led matrix of 40X60 in a wall. On arduino Nano, it gets out of memory. In the ESP32 we have more space for this. Have you tried with that big led matrix? Do you thing that I could organize the code to run with the led strips upside-down distribution (I have a hardware limitation)?
40x60 should be no problem for the ESP. Ignoring all of your other code, the LED array would need 7.2kB of ram. The ESP has 512kB, vs just 2kB for the nano! My code for this is designed to work in any orientation. There are some notes at the top of the code about using smart matrix to set the orientation and direction in which the LEDs are wired. Have a look at the adafeuit page for more info.
@@ScottMarley Thank you for the info! Another issue is that following the coments in the code, the led matrix width should be multiple of 16 and I would like to have 40. If I program 48 but only got leds for 40 I guess that I will lose the last 8 frequency bins, but the leds will still work. It’s correct? Do you have a better suggestion to get 40 width matrix?
How would this be modified to simply use usb power as opposed to a battery? Can you post a simplified circuit? Can you also advise where to find code that can use wifi to update the display to simply show an image? Perhaps a web interface that allows drag and drop to update the display?
To use just USB power, I would have the usb cable run into the ESP32 usb port, then run the power from the ESP32 5V pin to the LEDs. However, the ESP has a protection diode onboard that is only rated for 1A, so make sure you don't draw more than that (use FastLEDs set_max_power_in_volts_and_milliamps function). If you want to be a bit naughty, you can desolder this diode from the board and bridge across it with some wire or solder. This way I have drawn 2A through the 5V pin with no issues before. The larger display can run on 2A (although not at max brightness) so doing it this way you'd have to remove and bridge the diode.
Making this battery powered is a bizarre idea. Why would you ever want to do that? It's a VU, it's a stationary thing by its very nature. Also, why not build this around one of those glorious little LCD/ESP32 combos, like the T-Display S3? It might be alot smaller than an LED array, but it would look better on a 1.9" display, allowing for a more detailed, high-rez image that you could still appreciate from a few feet away.
Because my friend uses it as part of his mobile DJ setup, its not meant to be a scientifically accurate display, it's supposed to look good in a dark venue next to his kit. Which it does.
I tried to run this on this Model NODEMCU-32 and it did not work. Microphone test worked. Did I used the wrong pins? I tried severel conections. But maybe not the right ones.
Hello Scott! I've been researching and testing a lot of BMSs in the last 3 weeks, and I found this problem on almost all of them ( commercial battery packs BMSs and other DIY BMSs):They disconnect the output for 500 ms when the charge cable is connected or disconnected or both). But this doesn't happen on smartphones circuits, laptops, etc. So I suppose there must be some BMSs that don't do that. Do this two BMSs you are using behave also that way? I need one for a UPS with USB C and I couldn't find one yet.
I just tried both, and neither disconnect on plugging in USB power. I wouldn't recommend these though, I haven't tested them properly but it don't have much faith in their protection features!
@@ScottMarley Thank you Scott. Maybe they reset the output only when you disconnect the USB power. Most of them do that, and just some of them do it on connect.
Hello. Our projects are similar. I have a problem maybe you can help. I have a 5v power supply that feeds the leds. The microprocessor is also fed from this power supply. When I plug in the USB to run the program and start the device, everything works fine. But when I disconnect the USB, the LEDs turn on randomly as if there is noise. I tried with different power sources but the result did not change. Have a comment?
@@ScottMarley I used one side of the breadboard for the 3.3v and GND I got from the microcontroller. I made the microphone connection here. On the other side, I used it to distribute 5v from the adapter. The 5v/GND inputs of the microprocessor and the 5v inputs of the we2812b are connected here. I'm making a simple mistake but I couldn't find where it is..
@@aGGreSSiv that sounds fine to me. Have you used a multimeter to check that everything is connected as it should be? Check everything for continuity. On some breadboards the power rails are split in the center. Is that the case with yours?
That module is a sound threshold sensor. You set the level using the trimpot on the board, then when the sound is over that level the 'out' pin is set to high. If it's below that level it's set to low. That's it, couldn't be easier :)
Hi Scott, really liked this nice project. With your great explanation, I could easily set it up for a test run and it works great. But I have hard time finding out how to change the frequency range. Here it is set from 60 Hz to 5120 Hz with a sampling rate of 10240 and 512 samples. Trying your other project very helpful FFT Calculator Excel file, I can't find a relation between the Andrew's updated mapping code in "audio_reactive.h" and the results to modify his mapping for a new range. Would greatly appreciate a little help finding out how to change the frequency range for this project, if possible at all.
Hi John, sorry it took so long to get back to you. I've made a new version of the Excel file specifically for this project and uploaded it on GitHub. Have a look and let me know if it helps. You could change the sample rate to 20480 for example, and the highest frequency to 10240Hz (nyquist theorem says max f is half our sample rate). I haven't tried to see how fast I can sample with that microphone, but I'm guessing it's much faster than I am doing here! Let me know how you get on, I'd be interested to know.
@@ScottMarley Thanks a lot and hats off for your kind help and reply Scott. I've tried to get a grip on this new calculation, modifying the code for new ranges like 80 to 8000Hz, but testing it through my signal generator attached to an amplifier, the frequency distribution stays off :( Changing the Frequency Multiplier calculation From =(B7/B6)^(1/16) To =(B7/B6)^(1/15), shows the correct ending frequency on the generated table. But applying modifications in the code, it goes off and on higher than 5kHz, it rolls over backward. Tryied different Sample Rate, Samples or Ranges. Checking INMP441 Datasheet, it can respond easily from 80 to 10kHz with nearly flat response. Will setup another test bench later and try to read more, checking other codes, perhaps it finally becomes clear how to distribute a frequency range through 16 channels. Don't know why I have a feeling something may be missing in Andrew's update explanation to achieve this goal. Thanks again for your great project. As it is, it works great, respond precisely to music played and looks fantastic though :)
@@johnknight9422 that's ... interesting behaviour! Honestly I haven't looked into Andrews code too much tbh, and I'm too busy at the moment to give it much time. There's probably something simple that neither of us have though of, so if you do make progress on this I'd love to know about it! That way I can incorporate it into my code for everyone else to use. The other thing I wanted to integrate and never got around to is some kind of auto calibration. My idea was that you press 'calibrate' on the webapp, then play pink noise through the speaker you are using it with. It could then automatically normalise the heights of the bars as pink noise is supposed to be equal power in each octave (or something like that). It would be an interesting experiment if nothing else.
@@johnknight9422 I'm glad found this in the comment section. I was trying out the same thing and couldn't get an accurate representation of freq graph. I'm using a 8x8 matrix and found out it's only reactive in the low and mids section, barely get any treble. Am I missing something? If you find out something before me, would like to try out your findings.
@@ScottMarley I can't thank you enough for the projects that you have made and shared. They are just brilliant and the explanation is top notch. Thanks for all of it and keep them coming, has helped me learn a lot. Brightens up my space ;)
Yeah, the little one works well enough although I'm only averaging about 1A in the 8x8. The board in the larger one is definitely kicking out 2A, but I'd be wary of going near 3A. I think a better solution for the big one would be separate boards for charging and boosting, that way you can get much better performance and put the switch between the battery output on the charge board and the boost input to stop the battery discharging whilst it's turned off.
Hi Scott (and Andrew T), Thank you for a the great tutorial, and great code/ Libraries. I've been building it today, mine is going on the shelves in my lounge. I have hit a problem, I have 10 shelves, but your code breaks if the M_WIDTH is anything other than 16 (or perhaps 8). I've hacked the main sketch and that's now compiling, but hte FFT code looks like it is a bit harder to adjust to any number of M_WIDTH. If you have any way of helping with that it would be great. Drew
@@drewb3038 ah ok, the original code was never meant to work for anything other than 8 or multiples of 16. The reason is that o would have to dynamically change the number of bands of the FFT, and thats not a simple thing to code. Glad you got it working for your use case though.
Hi. Have you ever had the issue where your ESP32 won't connect to wifi when using external power supply? im using the "doit esp32 Devkit V1". the Wifi works fine as long the board is connected to my PC. But if I power it from a phone charger, it sometimes works and other times not. its seems a bit random. But when powering the board from "VIN" with 5V , the Wifi NEVER works. Have you had the same issue? if so, I will be very greatfull if you could explain how you fixed it. Thank you for creating the amazing content you are sharing here on youtube.
Hi Max, I haven't had that issue, sorry I can't help! How clean is your 5V power in? Is it from a decent supply or something really noisy? I can see that causing an issue.
@@ScottMarley Hi. im using a power supply from apple. When i google this issue, it seemes im not the only one that have this issue. some people says that its a mistake by the designer of the board. Do you own the exact same board? "doit esp32 devkit V1". i got mine from banggood.
@@maxbursell3513 mine looks almost identical to that, but it isn't 'doit' branded. I got mine from Amazon but there are so many different variants out there it's hard to know what I've got! I guess all you could do is try a differen ESP variant and see if that changes things. Good job they're cheap!
Hi Scott, my build is coming along nicely, but hit a roadblock. My INMP441 microphone is not being recognized. I am getting "Digital microphone is NOT present." in the serial monitor. I have tried two separate microphones and two separate ESP32 boards. I have tried changing the digital pin assignments. Using an ESP32 wroom 32. Any suggestions? edit: when I disconnected L/R from GND on the INMP441, the microphone was recognized. I understand that grounding the L/R pin causes the LEFT channel only o be used. I resolved the issue by changing .channel_format = I2S_CHANNEL_FMT_ONLY_LEFT, to "RIGHT" and left the L/R pin unconnected.
That's interesting to note. I'm not sure why that would have happened but thank you for posting the solution in case it helps somebody else. I'm using it on the left channel to maintain compatibility with sound reactive WLED but if it works better in the right channel then that's ok!
@@qualar I just buy them for free next day delivery from Amazon Prime, but that is a little bit more expensive and depends on where you live. I think from amazon they're about £7, not too bad to have it next day!
You must set the board to ESP32 in the IDE before compiling. If you've done that it should find i2s.h automatically. It's part of the ESP Arduino core.
![โกดำ - วสันต์17 x ไม้เมือง [Official Music]](http://i.ytimg.com/vi/OUg2L4Sx1TY/mqdefault.jpg)
If anyone is contemplating using the D1 mini form of the ESP32, you'll need to change some of the pin assignments in audio_reactive.h before compiling. I used the following (also mentioned on the webpage for the WLED fork):
#define I2S_WS 26 // aka LRCL
#define I2S_SD 18 // aka DOUT
#define I2S_SCK 22 // aka BCLK
Scott: trivial point but the web links for a couple of the required libraries at the top of your sketch are swapped around.
Hi, thanks for the useful information and the note about the libraries, I'll fix that later! I'll pin your comment for now so people can see it.
Michael, you gunna make a video of whatever yours turns out to be?
Good to see a man after my own heart - using spare network cable as a plentiful supply of wire. Neat and tidy project. Pro-tip - broken flat screen TVs are a good supply of diffuser sheets.
Brilliant work. I got a bit silly and made one 48 wide by 16 high. All because as a kid there was always a red led spectrum analyser in a shop window on Oxford street. I was fascinated by it and how it worked. Always wanted to build one and this project gave me the inspiration. I bet that one on Oxford street cost an absolute fortune back in the late 80s.
Really nice! The way you display the IP address on boot up is just brilliant
Thanks, for my use case it seemed the most sensible option!
Amazing, the perfect evolution of your first version 🔥🔥🔥
Always nice to see your videos. This one was amazingly well laid out. Congrats!
Cheers Andrew, thanks for the sound-reactive code!
Very slick Scott - Fast paced with lots of useful content. Battery power and mic with I2S interface.
Also loved the way you didn't stand on ceremony with the soldering - no PCB just flip the IC and solder the wires straight to the ESP32 pins.
Fit for purpose and gets the job done👍
Thanks for your comments :) for something like this a PCB would be more trouble than it's worth. It would be nice to have a custom PCB though that integrated everything with headers for the mic, and a charge and boost circuit all on board. But I'm not that good at electronics I'm afraid!
@@ScottMarley Nothing to apologise about. You've just illustrated that rather than getting hung up on jumping through hoops - pragmatism is king!👍
Also, I loved the way you side stepped the captive portal. You're solution for your purposes seems much better especially if other members of the family want to use it hassle free.
Personally, I feel like the captive portal is itself a fiddly workaround.
Why not use a cam enabled ESP32 and get it to read your network QR code.😊
Hey, someone has just made what I suggested
www.reddit.com/r/esp32/comments/lvlhl1/qr_code_for_esp32_web_server/?
OK - how about adding a time machine to it!?
This looks fantastic! It's such a beautiful, clean looking display and your patterns look so nice! I love how you display the IP Address on startup. That's a really smart decision. Thanks for the creating this vid with such clear and easy instructions, and also for the link to your code.
Glad you like it :)
as usual, a great video !
I discovered your channel a few weeks ago and only 1h after finishing the first video I ordered my first esp and led strip ! I now have a working prototype and i'm working on a box which can transmit signal to the VU display over wifi so that your sound source does not have to be close to the display (I'm not using a microphone so I only display the music and not the surrounding noise).
Thanks for the work you've done, can't wait for the next project !
Sounds interesting! Are you sending the actual audio data or doing it how WLED works and transmitting the results of the FFT? If the first way, is there any latency?
having just gone through the setup of my first couple of esp32's and doing a bit of playing, your videos have been such a help... to the extent my kids even said I was cool tonight. Im not telling them it was you but kudos and keep up with the vidoes. Thanks
Thank you, I love seeing comments like this! And don't worry, we're much cooler than the kids ;)
I was hoping to see something from you soon and what a lovely midweek treat.
That's very kind of you! I've got a bunch of things I'd like to do at the moment but lockdown in the UK along with too much 'real' work is making if difficult. Ah well, will get there eventually.
Thank you very much for your videos, I have collected your projects. Everything works very well. I modified the sketch a little and got very nice effects.
That's great to hear, thank you!
love this... great looking project....I have some spare esp32 and 8x8 matrices ...nice rainy day build.
Thank you, I was trying to keep it as simple as possible (but no simpler!)
very beatiful ! nice job, waiting for more stuff like this :)
151 / 5.000
Wow, what a wonderful video about FFT analysis with the ESP32! Your video really helped me understand the problem. Looks very well done!😃
Great Job Scott as usual. Thanks for sharing.
This one runs much faster than the other one you commented on as its using the second core to do all of the audio processing. It's a good idea to do this I think!
@@ScottMarley I think so too. I will try this one as well. Thanks!
Great UV Meter, love this project, I would like to make the 8x8 version, you did a good job Scott!
Thank you!
4:55 you mention you had to remove the usb socket, but didn't went into detail as to why. Great content as always Scott. Looking forward to your next project.
Thank you :) Because the USB socket would face the wrong way (into the side of the case) and I wouldn't be able to use it!
Nice one! Seems like a solid design.
I think I'll build it. I do a lot of pixel art, having a little frame like this running a server I can drop gifs onto sounds like a very fun project.
Yeah, that sounds great :)
Looks epic, nice work!.
Ha, cheers Matt :)
Great tutorial as always. Thanx a lot for sharing
To help make sure you don't run out of ideas :) How about a "nano leaf" project? For a little twist on the theme maybe some sound reactivity for activity around it.. ie some sound of a person in the room.. mellow animations.. some music maybe slightly more reactive and heavy metal.. very reactive.
It would be a very good topic for youre fastled Series where You explain how to use multiple groups and eventuely with different Animations or slitly different 👍🏻😍🔥
Hi Gabriel, what do you mean by groups? Are you talking about multiple strips connected to different controller pins, or groups of LEDs on the same strip or perhaps both? Either way, it's a good topic for a future video, and if I can find the time I might well do it!
@@ScottMarley Yeah, I mean both variations. What I am interested About is how You can animate lets say 5 groups in one strip simultane without or minimum delay. The Same About multiple Strips At different Pins. Thank You Scott!
Thanks for all the work you put into this awesome project. I put the 16x16 version together over the past few days and it came out great! I had many of the items on hand and only had to buy the display and microphone. Including the Fusion file was especially helpful as the acrylic sheet I had was not 2 mm thick. (Note that if you change the "Glass Thickness" parameter, you need to adjust the "Grid Thickness" accordingly). I did initially try 3mm translucent acrylic, but the display was blurry looking. I switched over to 2.5 mm clear with the printer paper behind it. The paper produces a little pulp pattern but the projected image is much sharper. Ultimately, I may try to find some thin translucent film to put behind the acrylic.
Great, so glad somebody else made one! You're right, there is a fault in the file as the glass and grid thickness should reference each other. I didn't notice this as I never changed that parameter to find out. Glad you managed to figure it out anyway!
PS do let me know if you discover a particularly good diffusion material, I'd love to find one that just works!
@@ScottMarley That seems to be more of an issue than I originally thought. I looked at a number of other papers that either have watermarks or the same fiber-ey pattern. Any white vinyl sheets I have found online are adhesive backed. At the moment, I'm using a piece of white UltraCote model plane covering material in the place of the paper. It's far from ideal. It likes to roll up, so it's really fiddly to get into place. It also has a heat-sensitive adhesive backing, so I'm not sure how it will work out in the long run.
I think I've settled on some Canon Glossy Photo Paper for the diffuser. A bit less light gets through, but the texture is very even.
@@karllaun2427 Ah, good idea! I've got some of that somewhere. Shiny side out I suppose?
Thank You for your work on projects, and sharing! One question / suggestion: would be very very nice to still have some physical button(s) controls option. I also put this suggestion on Github. I will try to add that myself, but, not sure will I succeed :) I am more hardware guy. Thanks!
Great content!
Amazing Project 😍
Thank you.
I love this! And your videos are great, thank you Scott! I'm having some trouble though I hope you can help with, I've connected everything up and everything seems fine, except there's just no reaction to any sound. If I touch the SD pin on the mic, the UV jumps on the matrix, so it seems they are able to communicate. I can't for the life of me work out how to downgrade to ESP32 core 1.0.6 and I can't find any info online - could you point me in the right direction? Or do you have any other suggestions? Thanks!
Scott, first I would like to congratulate you, you do amazing things.
I speak here from Brazil and I follow all your videos.
I built a matrix that worked perfect, but I can't get the same pattern from your video.
In this video pattern, the leds start to light up at the base and have this drop effect.
Is this pattern in your program available on github?
I'm not completely sure which pattern you mean, is it the one that's mainly blue? That uses the createWaterfall and moveWaterfall functions. Basically you color the bottom row depending on the intensity in each band (red highest intensity, blue lowest, createWaterfall) then every so often just move each row up (moveWaterfall). If you want to integrate it into your own code, you'll have to understand how mine works first, or it might be easier just to write your own version of it!
Awesome❤️❤️❤️.....also bro I said I ll try to make attachable PCBs for the led cube.....I could come up with the idea but making the pcbs Is beyond my level of expertise.
Well let me know if you do come up with a plan for doing it. I imagine it's harder than you might think it is!
@@ScottMarleySure😀.....and yeah it is😂
I've watched a lot of your videos and they are all brilliantly detailed. I'm building this project as my very first go at electronics, so I really appreciate how much care you have taken with this work. So (knowing nothing!) ... the code comment says 'LED data, D2 via a 470R resistor', but your wiring diagram doesn't show this. I'll follow the diagram :) Thanks Scott.
The resistor is optional but recommended. It will usually work fine without, but can refuse the chance of interference on the data line.
Use a simple Powerbank (like a 20000mah one) as the powersurce. It can also last very long.
I've done that on previous builds and it works fine and lasts forever! I wanted to investigate internal power this time. It sort of works, but I learned a lot doing it so that's more important!
@@ScottMarley yes thats true... I did also Use tp4056 charging Modules whith Protektion and an external Boostconverter for my Projekts. But i like to use cheap Powerbanks whithout Batterie, to recyce old li-ion cells from dead Laptop packs. Mostly only 1 or 2 out of 6 cells are dead.
For a 2s charging board i use the
"X-lite charging board" (its for a remote) it charges and Balancing the cells from 5v.
its a little extensive, But its the only Module i know that Balance Charge a 2s pack from 5v.
too bad u got subs under 10k, all of your videos is very great, clear explanation👌. I hope in 2021 u got more people that discover your channel bro
Hopefully one day. The electronics parts of TH-cam is fairly crowded, there are a lot of people with amazing content that I have to compete with but I'm trying!
Is it possible, like on the first version, to use double width bars? Thank you in advance.
Hi Scott! Thanks for your great videos! Any chance you can do some Fusion360 tutorials? Your teaching style is awesome so it’d be great! Thansk
Thanks for your comments! I'm not that good with Fusion at the moment so I don't feel too comfortable with making tutorials yet. There are plenty of really great channels out there that already do it really well though.
How do I increase the number of bands of this compared to the previous version
Awesome work, thanks for sharing
A question, but don't thing it's a small question lol.. Do you have a video on how to wire and program an SD card to ESP32, for example, to play music and lights effect files and pair it to multiple ESP boards through wifi, thanks
I'm afraid I don't. If it's just syncing light patterns up, you can use WLED which can sync multiple esps over wifi. Then perhaps have another on in charge of the music? You can get SD card readers with a built in audio output that talk serial to any microcontroller (I'm making one at the moment that runs from a nano). Maybe look into those devices?
Hi Scott, thanks for more amazing work. I think this is going to be my next project, but will likely use WS2812b strips (not the matrix). I would like 16 pixels wide but 24 pixels tall. Can you confirm that modifying M_HEIGHT to "24" will accomplish my objective? Thanks again!
It definitely should do, but I don't have any weirdly shaped matrices to test on unfortunately. If you get it working, feel free to let me know!
@@ScottMarley Hmmm. The matrix I built is 16 columns wide, each 20 LEDs high. I change M_HEIGHT to "20". For unknown reasons, the VU meter behaves like there are only 16 LEDs in height. Very strange. Still investigating how to convince that there are actually 20 LEDs in each column.
edit: Well, it turns out that I made a bad assumption. I am building this with strips of WS2182b lights, not a pre-built matrix. I assumed that the matrix was wired in a zigzag pattern starting from bottom left, going up, zigzagging back down, then repeating. I assumed wrong. The zigzag is horizontal. In the sketch, I changed "HORIZONTAL_ZIGZAG_MATRIX" to "VERTICAL_ZIGZAG_MATRIX". Issue is resolved. I also had to add 4 more zeros to uint8_t peak[ ], uint8_t prevFFTValue[ ] and uint8_t barHeights[ ]
Finally, one last thing: If you want to start your Data in (DIN) from the opposite side, and you explain the code, you can adjust M_WIDTH to a negative value, but please be mindful that it causes an issue with the formula for NUM_LEDS. That number cannot be a negative value, so replace the formula with the exact number of LED (corresponding to M_WIDTH * M_HEIGHT). Cheers!
Good day, thank you for the video. I repeated This project works very well, but I want to ask Cannot connect the button. If not, please add this feature, thanks a lot.
Parabéns! Perfeito!
It is amazing!! I liked it so much!😍 Scott, could you tell me how much all the printed parts of the case for a 16x16 matrix weigh in total?😀
138g standard PLA
@@Tavdog Thank you!☺️
Hi Scott,
while i verify the code it shows me an error, please help me fix out.
Compilation error: 'cLEDMatrix' does not name a type; did you mean 'LEDMatrix'?
hello friend, this project i want to use line in instead of microphone how do i wire and fix the code. thank you so much
Scott, I want to make a spinning persistence of vision (pov) display with fft. Have you thought about something like that? My vision for it would have an oscilloscope like output of the audio waveform in at least 3 and likely more bands. It would show only the peeks rather than a vu like output so you could see the trace of the waveform for each band. It would probably use bluetooth or wifi whichever I could get the lowest latency on. I'd want the whole controler plus LED strip to be spinning and power to get picked up through something like an inductive pickup so it could run all the time instead of having to get charged in a couple hours.
I think this matrix with fft is very close to at least the LED driving that I want to do. I'd like to have your mind thinking about this. What do you think?
Oh one last thought is that all the bands would have their moving peek of a different color sharing the strip so they'd be superimposed. The colors would have to combine if the peek appeared on the same LED. I was thinking say, 64 LEDs in the strip, or whatever the densest number one can fit in say 1/2 meter.
Mike
Good idea, are you thinking something like this create.arduino.cc/projecthub/theSTEMpedia/persistence-of-vision-pov-display-using-arduino-583d5f but with addressable LEDs? If the power is inductive, it would have to be a very powerful transmitter. For 64 LEDs you'd want at least 2A - even the best wireless phone chargers only just get near that and they're not moving. You lose a huge amount of efficiency if the coils aren't exactly lined up. I think you could get away with far fewer LEDs though, so might be doable with a wireless phone charger type arrangement. The patterns would be relatively easy compared to solving this issue I would think. I wonder how WiFi or Bluetooth would work when spinning round that fast?! I think it would be fine but would need investigating. The audio sync stuff over WiFi is already integrated with WLED, you could have a box somewhere with an ESP32 and a microphone sending the data over WiFi, and another ESP32 listening to the data and running the display. Just my initial thoughts, I do like the idea though and I might keep it in my head until I have some time for another project!
@@ScottMarley ok i do like the esp32 sending data over wifi idea, i also thought about slip ring power but didnt like the idea much, maybe i just need a big big battery? I have a windows media player visualization that does the fft and puts a scope in up to 6 bands. I've noticed even withing windows any lag really puts you off visually connecting the audio. So yeah I'd really worry about latency.
@@jmcarp0 I do think all this needs to be investigated further. You just made me order some wireless charging coils from AliExpress to experiment with ;) I won't have anything any time soon though, really busy with real life unfortunately.
Well I did order the esp32 2 pack and the microphone, and a 1 meter (144 LED) strip of ws2812b. I feel pretty confident I can get that going whether or not its spinning with big battery.@@ScottMarley
@@jmcarp0 good luck, and let me know how you get on.
Hey! Thank you for this project! I have a question. In my Arduino IDE, there is one error that keeps popping out: "Error compiling for ESP32 DEV Module", do you know how to solve it? It would help a lot!
Scott! Thank you for all the info! Really great project! I am trying to do a similar project with a big led matrix of 40X60 in a wall. On arduino Nano, it gets out of memory. In the ESP32 we have more space for this. Have you tried with that big led matrix? Do you thing that I could organize the code to run with the led strips upside-down distribution (I have a hardware limitation)?
40x60 should be no problem for the ESP. Ignoring all of your other code, the LED array would need 7.2kB of ram. The ESP has 512kB, vs just 2kB for the nano! My code for this is designed to work in any orientation. There are some notes at the top of the code about using smart matrix to set the orientation and direction in which the LEDs are wired. Have a look at the adafeuit page for more info.
@@ScottMarley Thank you for the info! Another issue is that following the coments in the code, the led matrix width should be multiple of 16 and I would like to have 40. If I program 48 but only got leds for 40 I guess that I will lose the last 8 frequency bins, but the leds will still work. It’s correct? Do you have a better suggestion to get 40 width matrix?
Thank you for sharing. I have two questions. Can I use a line input instead of a microphone? Can I use a button for mode transitions? Thanks
Yes and yes. I don't have any example code on this though, so you're on your own I'm afraid!
Great video.. shame the code no longer compiles
What’s the name of the boost board?
How would this be modified to simply use usb power as opposed to a battery? Can you post a simplified circuit?
Can you also advise where to find code that can use wifi to update the display to simply show an image? Perhaps a web interface that allows drag and drop to update the display?
To use just USB power, I would have the usb cable run into the ESP32 usb port, then run the power from the ESP32 5V pin to the LEDs. However, the ESP has a protection diode onboard that is only rated for 1A, so make sure you don't draw more than that (use FastLEDs set_max_power_in_volts_and_milliamps function). If you want to be a bit naughty, you can desolder this diode from the board and bridge across it with some wire or solder. This way I have drawn 2A through the 5V pin with no issues before. The larger display can run on 2A (although not at max brightness) so doing it this way you'd have to remove and bridge the diode.
Making this battery powered is a bizarre idea. Why would you ever want to do that? It's a VU, it's a stationary thing by its very nature.
Also, why not build this around one of those glorious little LCD/ESP32 combos, like the T-Display S3? It might be alot smaller than an LED array, but it would look better on a 1.9" display, allowing for a more detailed, high-rez image that you could still appreciate from a few feet away.
Because my friend uses it as part of his mobile DJ setup, its not meant to be a scientifically accurate display, it's supposed to look good in a dark venue next to his kit. Which it does.
I was working on music reactive led all connection coding done and now 39 leds are white and 40th is green and its not controllable
I tried to run this on this Model NODEMCU-32 and it did not work. Microphone test worked. Did I used the wrong pins? I tried severel conections. But maybe not the right ones.
find pin-out sheet here: drive.google.com/file/d/1Lw2Oz0USGsgaU1TTJEujdx9GcYBPtnDF/view?usp=sharing
Hello Scott! I've been researching and testing a lot of BMSs in the last 3 weeks, and I found this problem on almost all of them ( commercial battery packs BMSs and other DIY BMSs):They disconnect the output for 500 ms when the charge cable is connected or disconnected or both). But this doesn't happen on smartphones circuits, laptops, etc. So I suppose there must be some BMSs that don't do that. Do this two BMSs you are using behave also that way? I need one for a UPS with USB C and I couldn't find one yet.
I just tried both, and neither disconnect on plugging in USB power. I wouldn't recommend these though, I haven't tested them properly but it don't have much faith in their protection features!
@@ScottMarley Thank you Scott. Maybe they reset the output only when you disconnect the USB power. Most of them do that, and just some of them do it on connect.
hello i can't find driver/i2s.h library can you help me.
thank you
It is installed by default when you select an Esp32 board I think.
Can you share with us template for 3d print case for this project
Hello. Our projects are similar. I have a problem maybe you can help. I have a 5v power supply that feeds the leds. The microprocessor is also fed from this power supply. When I plug in the USB to run the program and start the device, everything works fine. But when I disconnect the USB, the LEDs turn on randomly as if there is noise. I tried with different power sources but the result did not change. Have a comment?
Not sure but it sounds like something isn't grounded properly. Either the micro or the LEDs. Other than that I can't think what would cause it.
@@ScottMarley I used one side of the breadboard for the 3.3v and GND I got from the microcontroller. I made the microphone connection here. On the other side, I used it to distribute 5v from the adapter. The 5v/GND inputs of the microprocessor and the 5v inputs of the we2812b are connected here. I'm making a simple mistake but I couldn't find where it is..
@@aGGreSSiv that sounds fine to me. Have you used a multimeter to check that everything is connected as it should be? Check everything for continuity. On some breadboards the power rails are split in the center. Is that the case with yours?
@@ScottMarley yes, the breadboard is the answer. I soldered everything together. No problem left. thank you for your suggestions..
how can I flip the display in the program. for me it's upside down because I start at the top right.
On line 74, try changing M_HEIGHT to -M_HEIGHT (put a minus in front of it). You might need to do the same with the width.
Can you show me how to use xd-74 sound sensor because it's to hard for me to learn
That module is a sound threshold sensor. You set the level using the trimpot on the board, then when the sound is over that level the 'out' pin is set to high. If it's below that level it's set to low. That's it, couldn't be easier :)
What's other alternative to make easier VU meter
Hi Scott, really liked this nice project.
With your great explanation, I could easily set it up for a test run and it works great.
But I have hard time finding out how to change the frequency range.
Here it is set from 60 Hz to 5120 Hz with a sampling rate of 10240 and 512 samples.
Trying your other project very helpful FFT Calculator Excel file, I can't find a relation between the Andrew's updated mapping code in "audio_reactive.h" and the results to modify his mapping for a new range.
Would greatly appreciate a little help finding out how to change the frequency range for this project, if possible at all.
Hi John, sorry it took so long to get back to you. I've made a new version of the Excel file specifically for this project and uploaded it on GitHub. Have a look and let me know if it helps. You could change the sample rate to 20480 for example, and the highest frequency to 10240Hz (nyquist theorem says max f is half our sample rate). I haven't tried to see how fast I can sample with that microphone, but I'm guessing it's much faster than I am doing here! Let me know how you get on, I'd be interested to know.
@@ScottMarley Thanks a lot and hats off for your kind help and reply Scott.
I've tried to get a grip on this new calculation, modifying the code for new ranges like 80 to 8000Hz, but testing it through my signal generator attached to an amplifier, the frequency distribution stays off :(
Changing the Frequency Multiplier calculation From =(B7/B6)^(1/16) To =(B7/B6)^(1/15), shows the correct ending frequency on the generated table. But applying modifications in the code, it goes off and on higher than 5kHz, it rolls over backward. Tryied different Sample Rate, Samples or Ranges.
Checking INMP441 Datasheet, it can respond easily from 80 to 10kHz with nearly flat response.
Will setup another test bench later and try to read more, checking other codes, perhaps it finally becomes clear how to distribute a frequency range through 16 channels.
Don't know why I have a feeling something may be missing in Andrew's update explanation to achieve this goal.
Thanks again for your great project. As it is, it works great, respond precisely to music played and looks fantastic though :)
@@johnknight9422 that's ... interesting behaviour! Honestly I haven't looked into Andrews code too much tbh, and I'm too busy at the moment to give it much time. There's probably something simple that neither of us have though of, so if you do make progress on this I'd love to know about it! That way I can incorporate it into my code for everyone else to use. The other thing I wanted to integrate and never got around to is some kind of auto calibration. My idea was that you press 'calibrate' on the webapp, then play pink noise through the speaker you are using it with. It could then automatically normalise the heights of the bars as pink noise is supposed to be equal power in each octave (or something like that). It would be an interesting experiment if nothing else.
@@johnknight9422 I'm glad found this in the comment section. I was trying out the same thing and couldn't get an accurate representation of freq graph. I'm using a 8x8 matrix and found out it's only reactive in the low and mids section, barely get any treble. Am I missing something? If you find out something before me, would like to try out your findings.
@@ScottMarley I can't thank you enough for the projects that you have made and shared. They are just brilliant and the explanation is top notch. Thanks for all of it and keep them coming, has helped me learn a lot. Brightens up my space ;)
So is that charge/boost circuit able to output 2A through the boost converter?
Yeah, the little one works well enough although I'm only averaging about 1A in the 8x8. The board in the larger one is definitely kicking out 2A, but I'd be wary of going near 3A. I think a better solution for the big one would be separate boards for charging and boosting, that way you can get much better performance and put the switch between the battery output on the charge board and the boost input to stop the battery discharging whilst it's turned off.
Hi Scott (and Andrew T), Thank you for a the great tutorial, and great code/ Libraries. I've been building it today, mine is going on the shelves in my lounge. I have hit a problem, I have 10 shelves, but your code breaks if the M_WIDTH is anything other than 16 (or perhaps 8). I've hacked the main sketch and that's now compiling, but hte FFT code looks like it is a bit harder to adjust to any number of M_WIDTH. If you have any way of helping with that it would be great. Drew
That's odd, it should work for any multiples of 16, but I haven't got a matrix to test that on. What width are you using?
Well, I've half bodged the code. But I think (hope) I've now got it working for 10 columns.
@@drewb3038 ah ok, the original code was never meant to work for anything other than 8 or multiples of 16. The reason is that o would have to dynamically change the number of bands of the FFT, and thats not a simple thing to code. Glad you got it working for your use case though.
Hi. Have you ever had the issue where your ESP32 won't connect to wifi when using external power supply? im using the "doit esp32 Devkit V1". the Wifi works fine as long the board is connected to my PC. But if I power it from a phone charger, it sometimes works and other times not. its seems a bit random. But when powering the board from "VIN" with 5V , the Wifi NEVER works. Have you had the same issue? if so, I will be very greatfull if you could explain how you fixed it. Thank you for creating the amazing content you are sharing here on youtube.
Hi Max, I haven't had that issue, sorry I can't help! How clean is your 5V power in? Is it from a decent supply or something really noisy? I can see that causing an issue.
@@ScottMarley Hi. im using a power supply from apple. When i google this issue, it seemes im not the only one that have this issue. some people says that its a mistake by the designer of the board. Do you own the exact same board? "doit esp32 devkit V1". i got mine from banggood.
@@maxbursell3513 mine looks almost identical to that, but it isn't 'doit' branded. I got mine from Amazon but there are so many different variants out there it's hard to know what I've got! I guess all you could do is try a differen ESP variant and see if that changes things. Good job they're cheap!
Can I use any WS2812B 16×16 led matrix
I don't see why not.
@@ScottMarley after that can I change it into animation screen
@@abhijithnscs1364 you can do whatever you like, but unless you know how to code that you'll have to find another project which displays animations.
👍👍😍
Hi Scott, my build is coming along nicely, but hit a roadblock. My INMP441 microphone is not being recognized. I am getting "Digital microphone is NOT present." in the serial monitor. I have tried two separate microphones and two separate ESP32 boards. I have tried changing the digital pin assignments. Using an ESP32 wroom 32. Any suggestions?
edit: when I disconnected L/R from GND on the INMP441, the microphone was recognized. I understand that grounding the L/R pin causes the LEFT channel only o be used. I resolved the issue by changing .channel_format = I2S_CHANNEL_FMT_ONLY_LEFT, to "RIGHT" and left the L/R pin unconnected.
That's interesting to note. I'm not sure why that would have happened but thank you for posting the solution in case it helps somebody else. I'm using it on the left channel to maintain compatibility with sound reactive WLED but if it works better in the right channel then that's ok!
Thank you so much for writing about your fixes, it helped me a lot!
anybody that can recommend how to make it so, that you get both analog in and microphone, and then able to switch between them?
I cannot for the life of me get the microphone working in WLED. Can anyone tell me how I can confirm if the mic is working?
Are you definitely using the sound-reactive fork of WLED and not the normal one?
@@ScottMarley 100%. I think i have a faulty mic. I tried the test sketch and that doesn't pick anything up on the mic either.
Ah, that's the next thing I was going to suggest. Does sound like a buggered mic then. Ah well, least they're fairly cheap ;)
@@ScottMarley Yeah its just waiting for the things to arrive that is the pain.
@@qualar I just buy them for free next day delivery from Amazon Prime, but that is a little bit more expensive and depends on where you live. I think from amazon they're about £7, not too bad to have it next day!
El micrófono se me ha dañado 2 veces
3d stl product how to buy
STL files at the GitHub link in the description
+++++
Good job. Could you put a link to the battery charger you have used. Specifically, the one you used for the 16x16 matrix. Thank you.
I wouldn't recommend the one I used for the 16x16 as it couldn't cope very well as it turns out. You'll need a 3A minimum buck boost lipo charger.
driver/i2s.h is missing could you help me?
You must set the board to ESP32 in the IDE before compiling. If you've done that it should find i2s.h automatically. It's part of the ESP Arduino core.
@@ScottMarley thank you for your immediate replay