Loved it! As a drama you'll demonstration sounded way more musically interesting than the demo. There is not much of a substitute for an actual human triggering whatever is making the sound. This video really helped me realize some of the things I need to think about in order to tape one of these things onto my bass drum and have it play crazy sounds whenever I flick a switch I'm going nuts with double kick. Thanks. :-)
One of the coolest things I've seen on RU-vid in a long while. I've actually been thinking of modifying an existing eDrum set to essentially connect via Ethernet (instead of 1/4" jack) and completely bypass the sound module or a drum trigger interface and connect to a generic 8 port switch > into MacBook and trigger Drum Sounds within the software using Network MIDI. Spawning some ideas.
Awesome video thank you! 😅👍 (If you want to beef this device up, you could configure the Teensy to assign MIDI note numbers to each of the triggers and emit MIDI note events via a 5 pin DIN jack.)
I still need to make that PT2399 delay/reverb thing and a million other things. In time! I guess these days when I say stay tuned, that has a tolerance of up to 1 year!
Great video. I used these piezos to make triggers to put in acoustic drums to drive LED strips - one trigger for each drum (inside the drums - trigger and LED's), using an STM32F1. It was pretty easy to catch only the initial peak as you have noticed. The problem was that they were not road worthy - they would literally fall apart after at most 8 or 10 shows - sometimes 2. I eventually embedded them inside a coating of liquid electric tape and that worked until not much longer when the thrill of drums that light when you hit them wore off. I was thinking it could be cool to a web sort of thing above the drummer sort of reflecting the drums played, but I haven't picked the project back up.
I was thinking of eventually trying to make drum triggers so "playing" can be recorded and then the kit can be changed for playback for fun, but in reality what I'd need is a circuit that can do the equivalent of auto-tune for bad rhythm....need lessons. But I have an acoustic jellybean kit made of Tama Rockstar (main 5 piece plus an add on 10 rockstar tom), Yamaha Stage Custom (8 tom), Pearl Export (18 floor), and a couple of 6 and 8 Tama Artstar (I think) concert toms. 20 years of living next to a music store and having access to ebay.....
This peak detector is way better than my first attempt at a drum module! It resolves a lot of the problems I was having. The bad part is the unity gain. Once the piezo is installed in my diy mesh head or glued to a cymbal, the voltage drops considerably. My home made edrum set is pretty typical cone trigger and glued cymbal attachment. I've been experimenting with modifying the circuit to get me closer to a full range of velocity (0-3.3v). The unmodified board gets me around a half volt from the drum and about 20 mv for the cymbal. After tacking on a voltage divider and diode for the gain, I get about 3v for the drum and 1v for the cymbal when I wack it my hardest, but not consistently. Hopefully I'll know more when my new oscilloscope gets delivered. If you do revisit this circuit, I'd love to see how you would go about getting this to work. Thanks a bunch. My first attempt had lousy, inconsistent velocities and couldn't even detect a cymbal. I can't begin to tell you how much of an improvement this has made. I've learned a great deal.
I wonder if some sort of compressor op amp circuit would help where it can amplify weaker signals and attenuate stronger signals automatically calibrating but of course the responsiveness would need to be looked at. I’m sure eventually I will be looking at trying to trigger acoustic drums with it and I may try to figure out how to normalize the conditioned signals then.
OOOh this is nicely done. Once I saw video where they used Arduino nano with some multiplex(i think, it was long ago) and MIDI conventor. But it was kinda messy. This looks way better and more responsive.I really want to build it :D Nice work and nice video ! :)
I might dabble with MIDI eventually but I like self contained things more. The responsiveness took a bunch of trial and error both in hardware and software to be able to pick up fast sounds, reject false triggers, and respond to the intensity. Most projects I see just detect on/off and play something.
@@GadgetReboot If i remeber correctly, he could sample signal strength from the piezos. But it worked more like MIDI keys and then the FL studio plugin did everything(like leng of the sound, volume and so on). Thanks for good video and nice ideas :)
Sadly, I could have had a rockband drum set for free this past year when helping someone clean out before they moved, and I used false logic of "I'll never have a use for this electricity related item"
@@GadgetRebootI know this is quite an old video but with the Hero/Rockband franchise dying out Clone Hero instruments are being made. I am working on an old electric drum kit that was fried with lightning so the pads are still good but the drum brain is shot. I was following this tutorial and was wondering if the circuit needs to be modified for long piezo wires. The kits drumheads have a voltage divider looking circuit with a 1MOhm across the cathode and anode wires, I am guessing this is what you have in your design. There is also a 51KOhm resistor on the cathode wire that sits in series between the piezo element and a T junction with that 1MOhm wire. I think this is to drop the voltage a little or "tame the hotness" as electric drummers like to say. Would your design have to be modified to work with this setup? I doubt I'll get a response but happy to get one nonetheless.
Nice video, i've got a few piezo transducers laying about somewhere. most of the components and a Teensy 4.1 with a teensy audio board. looks like i've got to make a drum kit :) Need to play with the Teensy a bit more. Thanks for something to play with. By the way still using the arduino fuse setting board, used it just today to reset the fuses of an ATTiny85
I have my fuse board right here but the OLED is nowhere to be found - I need to make one with a soldered display copy too. I had to severely cut this video down so a bunch of details didn't make it through, and I also made notes for a separate video regarding obtaining and using audio samples in Teensy program memory, converting wav files to data arrays and some gotcha's I learned. The original video would have been an hour long...
I had extra video clips left over along with screen capture pictures and sound files that I knew were supposed to be in the video somewhere....but couldn't figure out where. Then I was missing some files, re-recorded voice overs, then found the original audio in another folder. This is what happens when it takes multiple sessions!
This was absolutely awesome and you really gave a great explanation. I think the circuit would have been a lot simpler though, if you had just used a Quad 555 timer Set up as monostable multivibrators triggered by the input. I think in that case, you may only have needed the zener.. Or maybe a couple of in914's, and use the voltage drop across them. Great video, and I'm definitely a sub.
Yeah I could have triggered a one shot like that as a sort of debounce, but I'd still want to have a way to detect the strength of the hit so I'd have to see if I could trigger it enough to generate longer/shorter effective pulses that can be time measured as an intensity indicator. I haven't played with the monostable in a while. If the input trigger is still in a trigger state when the output has finished its pulse, if it can re-trigger immediately without needing the input to cycle first, that could work. But of course I also wanted the interface to potentially be used in further analog noise making circuits to trigger those with analog inputs, either to modulate their volume or pitch based on triggers. It's good to have many ways to do things. I'm going to try to get back to these types of building block analog circuits this year.
This is the light at the end of the tunnel holy grail for me !!. Lots of the heavy lifting hard ware science resolved .I have to connect this to 32 piezos and my max patch and some sketch miracle}; , somehow seems attainable now .My crash course into free tech world could soon be completed (6 months estimated, and I can stop patching and Arduino and go chop wood. Thanks sorry didn't study the whole thing before I commented. Im going to be here for a while Ill be buying a Buch of those condition boards soon..
if there were no obvious discharge path like a resistor so that the capacitor can only either see a diode or the input of an op amp, there is still a very small amount of leakage current in reverse through the diode and possibly into the op amp, whether it’s microamps or nano amps, so eventually it would discharge, especially being a relatively small capacitor value as opposed to something like a big electrolytic that still may hold a charge for a very long time.
I think I was using three different I2S boards and chips interchangeably with no problem but it’s been a while since I’ve used any of them. The main difference might be that the SGTL 5000 might need some different initialization code but the hardware should all be able to handle the audio data
This has got to be the best video regarding DIY edrum ever, very informative and easy to follow. Thank you very much for sharing this with us. I have a question, Can this work with older version of the Teensy board? The 4.1 version is not available where I live :(
thanks! It should work with previous teensy boards, I’m not sure if there might be slight changes in how they store files in memory but I know they could do it before the version 4 boards.
@@GadgetReboot Hi! do you thin it's possible to add something like this to the teensy board? ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-dbzGuadzXY8.html
Hi! Very nice video, thanks a lot for that! I have a question: I haven't tried out the circuitry but just having a quick look makes me wonder if the Zener doesn’t limit the dynamic range of the interface significantly. For example, my edrum pads output up to 30 Vpp, with the positive component (after rectification) being roughly 20 V, if I hit them hard. Here the Zener would result in all hits giving more than 5.1 V piezo output being clipped to the same value, wouldn’t it? Thus, I would lose ¾ of the dynamic range. Or do I misunderstand something here? Would it be a workaround to use a 100k trimmer in the input that would still load the piezo with 100k but also giving the possibility to scale down the input to the full dynamic range of the drum pad?
are those pads pretty much the same thing, just a striking surface with the output of the sensor and no other electronics until they plug into something else? The reason it’s scaled down to 5 V is because that’s the limit of the digital circuit it ends up plugging into so whatever those pads plug into probably ends up scaling that down as well to a safe level. I don’t know if it works out directly proportional but it’s possible some dynamic range is still preserved by clamping it at 5 V max because it’s sitting at 5 V for the duration that the sensor itself would be higher so the amount of time it’s at 5 V may be the translated dynamic range of how high the signal managed to get. So if it only peaked at 10 V it may drop down below 5 V sooner and have a shorter pulse width but if it got up to 30 V it may stay locked at 5 V longer and however it’s handled in software, a longer 5v pulse can be interpreted as a louder or harder hit.
@@GadgetReboot Sorry for my delayed response! Yes, I unerstand the reason for the 5V limit. My question was just concerning the fact that the input seems not be scaled down but actually clipped. However, I didn't realise that the time course of the (clipped) signal might still contain volume information. I will start experimenting a little bit and then decide what might fit my needs. Thanks a lot for your explanation!
thankyou for sharing. I am yet to dive in and buy the kit. I want to make a midi djembe (because my ears cannot take real drums and my partners hands cant take them either). The piezo signal altering was very interesting. I used to be a programmer but electronics like that leaves me baffled ! lol i can read code but not circuits...... it will be different for hand drumming piezo response but this hellps a lot. When you triggered play and then a few ms later before full decay you play again for the next trigger, did the entire sample of the first get played? or does the second trigger stop the sample playing and start it again? ie does the sample play time ever exceed the gap between hits? i ask because it sounded good on the demo. I am also interested in after touch or continuous touch.... since with a djembe if you play the right hand with the left hand on the skin, the sound from the right hit is very different. thanks again
I glanced at the code it’s been a year now, so it looks like if the same instrument input is triggered again, it will cancel and start the sample again but when I’m playing multiple separate instruments they play concurrently. I think when I set it up, I was thinking that for a regular drum kit, when the stick hits it again, it’s overpowering The original hit on that same instrument anyway so I might as well just take the easy way and let it start the sample from the beginning but of course I want multiple different inputs to be going at the same time. I think the way it works is each sample is assigned to a sample player in memory so with eight inputs I have eight players sitting in memory and all eight can go at the same time, so if I actually wanted to be able to re-trigger one of them multiple times and have those all play simultaneously, I would probably duplicate or triple the number of players in memory and cycle through them so I actually get three concurrent hits on each Piezo input or however many I think I need to be realistic with whatever is being played by hand. But I’m not sure if that eats up a whole lot more memory to actually load the same sample over and over, or if there’s more clever ways to do it like just redirecting the existing sample to different players, it must work that way I can’t imagine they set it up so inefficiently in the audio libraries, three separate players must be able to read from the one data sample. but that’s the concept. to simulate something where hitting it different ways may actually require a different sound sample playing back instead of just the same sample at a different volume, which I also imagine a bodhran drum benefitting from, if it’s being simulated with maybe two different piezos on the same surface, I suppose software could look at the inputs as a set and do things like if only one input is triggered, just play a standard drum sample at whatever intensity it was hit, but if the other input also triggers relatively close, maybe assume the skin is being muted and playback a whole different sample at whatever intensity. but if the hand is constantly resting near the one sensor it may require a different type of input hardware monitoring for pressure sensitivity while the current circuit is more looking for momentary spikes instead of constant pressure but it’s definitely doable
@@GadgetReboot thanks for the reply yes as for the second sensor i guess an FSR (force sensitive resistor) would do, but i saw one attempt at building a midi drum with them instead of piezo's and they were erratic. As for how the output of a quickly hit piezo behaves if the hand stays on i dont know. Are peizo's constant pressure sensitive and the oscillations in the decay voltage created by physical ocillations in the sensor pad?? The use of the metal base might signify that. Stick heads are very small of course by comparison and may preference decay oscillations in a hit. (thus the decay capacitor might mask the nuance of such hits. I dont have an oscilliscope but that serial output function will help. Yes ironically a djembe similator would have to accomodate both situations ... the sound of a strike is cut off by ANY new hit, .... but also changed by a constant pressure on a pad (same or different pad). The bass on a djembe for example is immediately muted by a subsequent tone note because it is on the same skin. However it isnt an immediate cut off, and rim hits have less sustain effect on previous bass strikes and other notes. On the other hand (excuse the pun) fingers or palm on the skin have a distinctive altered sound on the strike note. (even to the point of creating a sliding scale as the finger is moved to the rim). I wouldnt even try to emulate that! But in addition, there is the delayed hand release (of flap), which also mutes and changes the sustain and sound of a note, and djembe drummers use that instinctively. All very complex! lol ... but thanks again for your very detailed reply. Even the sounds you got were very good and i cant expect a midi peizo to be the same as a single skin acoustic. With regard to the opposite desired effect(i am not familiar with sticks) it would be more like a polyphonic sample duplication like you say. Quite how of course i dont know, but luckily it isnt what i am after for a djembe pad. (multiple mixers i the gui?? Can a sample be played by 2 mixers at different times but within the same note length on a teensy?) anyway start simple! 👍
Hi! First of all thank you very much for putting the effort into making this video. It's great! I'm trying to use the same design to build a diy edrum kit but I noticed that after adding the Zener diode the signal gets clamped to a fixed value. So the initial part of the decaying signal is flat at the Zener voltage for a couple of ms and then, after some time, decays normally. This makes it hard for the ADC to distinguish between hits stronger than a certain threshold. While this is expected behavior with a Zener voltage regulator, it doesn't seem to happen to you in the video. Am I missing some detail? Thanks again!
I can’t remember now what my waveforms and voltage levels looked like but my first guess if the signal is staying clamped at the zener voltage too long, it sounds like maybe the piezo voltage is staying high for too long so if it’s above 5 V for a couple of milliseconds it will stay at 5 V if that’s the Zener used. if you are using the same component values and circuit that I am, I wonder if the capacitance on your piezo is larger than mine so it’s holding the charge longer. If that’s the case maybe reducing the resistor value that’s connected to the piezo to discharge it faster would help quickly bring the voltage below the zener threshold.
can't thank you enough for this. i'm working on a desktop sized trigger with the same layout as a certain javanese instrument to use as a practice tool. the instrument is comprised of two rows of seven horizontal gongs. the scale ascends left to right on the lower row and ascends right to left on the upper row an octave above. so it's a bit of a brain breaker if you're not used to it. it's also enormous and i wouldn't even know where to get one if i had the money and the space! so how to practice? not using MIDI hadn't even crossed my mind! but a self contained unit that just does the one thing it's meant for has an elegance to it. some serious food for thought. of course i would either need two 7 channel boards or 4 X 4 to pull it off... anyway, thanks for the great explanations of each component's job in the circuit. this solves the issues i was facing with piezo and teensy. very excited to look more deeply at the code and see what ideas are born;)
Thanks! I still have your shipment of parts here to put to use - got off track this past year (and still counting) with moving but I will eventually use every item anyone ever sent me in a project! Maybe one big chaotic project with everything!
@@GadgetReboot no problem. I'm sure that I will never use the parts. I've got 39 formal projects in the queue now. and counting. I'm trying, really hard, to focus on implementing ROS into my robots. Steep learning path, but has some big upsides. PS: loved your work with the 555 and ATtiny85.
One reason for all of the signal conditioning was to help control the signal because once the sensor was struck, sometimes it would hold on to a voltage for an extended amount of time before the charge goes away so the ADC would see this extended duration maximum voltage based on what the Zener clamps it to. that was making it difficult to detect multiple rapid strikes. they would just look like one long single hit so the components help to discharge the transducer and then It can pick up another signal instead of it getting lost. and with all of those extra components, that’s why the op amp was needed to buffer all of that from the ADC, so that it doesn’t impact the behaviour of all of the charging and discharging timing.
I don't know what impact they can tolerate or if they're even spec'd to be used that way (without some buffer material) but the spudgers are light enough. I think in a real drum trigger scenario the piezo would be just picking up vibrations while mounted as an indirect pickup.
Hi, congratulations on your project, it seems really well done and above all well explained. Maybe I ask a question already asked: you built and calibrated this circuit by hitting the piezos directly. In a battery with mesh heads, the piezo receives much less stress than hitting it directly, I breadboarded your circuit but on very low hits without the op-amp receiving almost nothing to amplify, so its function is very limited, do you have any suggestions?
if there is still a signal but it is weak, I would probably be performing the same calibration playing with existing component values in a prototype until I get a signal strength that works with the set up or , adding another op amp stage at the output can boost it, as long as the maximum voltage is still going to be within safe limits for the input it goes on to.
Thanks for this very informative video, I have a couple of questions: 1) would this work with digital input as well (basic on/off - value not necessary)? 2) If I wanted to do this with only one piezo which single channel opamp would you recommend? Thanks again!
I use the LM358 op amp as a general purpose and there’s two of them in the one eight pin package. I would expect that to work fine here. I also would expect a digital input to work to trigger samples because the analog input is just an increasing and decreasing voltage that is within 5 V logic levels going into the op amp, so the digital zero to five or even zero to 3.3v would be the same as the analog signal jumping directly to those voltages instead of more gradually going up and down.
Thank you. Really nice and illustrative. I am making a drum kit, but I am using a 4067 MUX to feed 16 piezos into one arduino analog input. Is it possible to do the signal conditioning just once at output of 4067 or is it required to do it at each input?
I would say each input needs conditioning separately because otherwise all of the high voltage and negative voltage spikes will be presented to the 4067, which likely can’t tolerate it anymore than a micro controller input. And if all the different sensors are routed momentarily to the one signal conditioner, the point of the signal conditioner is also to hold onto the trigger pulse so it will be available when the processor can get to it so if there is a sensor that’s not currently connected to the conditioner, the pulse may be missed or at least its characteristics like what the peak was and how strong the hit was if the micro controller only gets to it halfway through the pulse when it’s already decayed.
@@GadgetReboot Very clear now. Thank you very much. What do you think about having all the conditioning for each piezo before the 4067, but only one OpAmp buffer after the 4067?
It can detect some pressure/weight if something is placed on it but I don't know the amount of minimum force needed to register. It probably wouldn't pick up light plastic but maybe a metal one.
@@GadgetReboot Thanks for the fast answer. So I will keep searching for another solution. The most used one is IR leds, yet I want something hidden under the tracks.
@@YigalBZ I wonder if this would work www.aliexpress.com/item/4000907686535.html there is some info on how to use it here innovatorsguru.com/strain-gauge-sensor/ but I can't find info about what type of force it should be used with or if there is such a thing as too much. but given its size, it looks like it would work with light loads. Though maybe the weight of a track would already be too much. Something to investigate. I might get one of these myself
@@GadgetReboot Thanks ! This can be very interesting. I will order few for testing. I am also going to try hall effect sensor, but this will require metal which I am not sure will work as a generic solution. Your idea seems to be better.
I'm curious to see how others have mounted it in an enclosure? It would be nice to mount it in a way that you can access the pots and and see the leds. Anyone have pics of their final build?
Hello.... I tried to understand everything, but since english is not native, i might have miss something. If I follow and build everything you show, i dont need a computer and softwares to play? It will work as a modulo as the commercial drums do?
if this project is duplicated, there is no computer needed after programming the teensy 4.1, all of the sounds are in there and it works as a standalone player.
super cool man. why use the zener and all that when you have the 1n4148 which drops the signal by .7v if the input is 5v. you probably said brother but i missed it. i get why you used the opamps. i like how you use them all the time, it shows your experience and the fact you want to keep the signal integrity. i suppose the dc input to the opamps are 0 on vss and 5 on vcc ehh? or could you put higher voltage through them then trim the output with the pots on the output? awesome circuit man.
The zener was used just to try to keep the signal no higher than 5v but let it pass unaffected within 0-5v because the peaks can be anywhere up to tens of volts, so I'm keeping it max 5.1. I don't remember the max input op amp voltage on that part relative to VCC but definitely not higher than VCC. I keep meaning to do a revision of the circuit but then I keep forgetting what changes I wanted to make.
Amazing explanation - Have used your sketch to build a digital Handpan and it works perfectly. Would it take much to pull the audio from the SD card instead of memory? Thinking about adding different scales in so will have to use an SD to hold them all.
I know the data can be pulled from the card easily enough, in the Teensy audio gui tool there are PlaySDWav and PlaySDRaw that can be used as sample sources, but I haven't tried it so I don't know if there is latency that would cause problems for playing back immediately. Wav especially would take time to parse the header each time etc, raw plays immediately which sounds like playing from program memory, just don't know if the SD vs program memory access time is too slow.
Yes I can't remember how I set it up for the real time playing but for the automatic demo playback I had maybe 4 channels and I would play the next instrument in a different channel to give the previous one(s) time to finish ringing out, or allow simultaneous hits like bass/snare or two toms. It's limited by memory and how many different internal players we can create and mix without distortion, but then it's just a bunch of files played from memory on the different players with the outputs being software combined.
@GadgetReboot im back on this project after being busy with work, im thinking to order this from PCBway, but before that, i wanted to ask if theres any improvement you make on the circuit design?
@GadgetReboot awesome! Pcbway review my order currently, im going to use it with my old teensy 3.2. For sure need to do change the io assignment..but else than that..do you think the code run on teensy 3.2? Thanks for the reply
The oldest board I have used is 3.6 so I’m not sure what the difference is, the only special thing the software does is load the audio samples into memory to play them back so maybe there could be a problem if that other board doesn’t have enough memory so you could try compiling the sketch for that board and see if it fits, and if there’s not enough memory maybe you could remove some audio samples or use smaller ones
@@GadgetReboot ive ordered the board from pcbway, id love to get your assistance in this if you provide the service, i have a very specific project that i want to establish.
i want this. with the sticky tape,wires all over the placeand everything, I like to make things see through or open. I do this so my children can see how all things are simple. We can make anything we want.....
It's been months since I did it but I think the 20ms was the time between successive hits being registered rather than latency from making a hit and hearing the sample played, and so it was more about the stick responsiveness ability to detect fast multiple hits. Somewhere after 12 minutes in, showing the waveform of 2 hits on the scope and talking about how the actual sound is triggered when the signal falls 5% below the peak, I don't remember the scope time per division but the signal is triggered quickly with the cleaned up trigger pulse, then the next stick hit is detected some ms later and relatively quickly registered to make a sound again. If I had everything set up better it would be interesting to actually measure with a microphone hearing the stick hitting the pad and hearing the sound start to play (plus the latency of physical spacing but the mic, speaker, and pad would all be within a foot), but maybe that can be for another day.
@@GadgetReboot OK so it sounds like latency shouldn't be an issue then. I am about to build this circuit to use with a 3D Printed 12" snare drum i've just made. Thanks for all the effort you put into creating this.
Those latencies are good for drums but other instruments because of different latency won't play that good unless we use designated piezos It could literally go into hundreds of thousands of variations based on what samples are used)
First, amazing video! I learned a ton just from the piezo signal conditioning segment alone! I have a question regarding your choice of the Teensy microcontroller. I was hoping to use an Arduino for my project, but I'm struggling to play multiple sounds simultaneously (for example, a kick and a hi-hat are often played on the same beat). Is the Teensy board capable of playing multiple sounds simultaneously? Also, would it be possible to write code so that, depending on the strength of the signal from the piezo, a different sound file is triggered? I'm thinking for example, light taps on the snare pad trigger a rim click sound, but stronger signals on the snare piezo trigger a full snare sound. My project will also be battery-operated, so I'm looking for a Microcontroller with a low power draw. When trying to choose a low-power microcontroller that can play/mix multiple sound files simultaneously, what sort of specs should be looking for? Thanks!
with the teensy it can play multiple different sound files simultaneously because they are all set up as if they are a separate sound source going through a mixer so the sound just gets triggered and then it plays through the mixer in software. I haven’t tried it on anything else so I don’t know if other software libraries on other chips can easily do the same but I just wanted to get up and running fast. software can look at the analog trigger input and determine if it is a soft or hard hit and then decide to play a different sound sample because the strength of the hit causes a different peak voltage so decisions can be made based down how strong the input is. The harder part might be calibrating it so that it consistently knows the difference on a single input because there could be a legitimate quiet hit intended for one type of sound to be triggered but it gets interpreted as too quiet of a sound and it triggers the other thing so multi zone separate triggers might be more ideal like how they do triggering on a snare with one dedicated sensor for the snare head and another dedicated sensor attached to the shell for rim hits. for power consumption I don’t know if it directly would relate to the ability of a chip or its software to play multiple files simultaneously, it would probably come down to how fast the processor runs and if it has more on board resources that need more power in general to run so if the project is just to blink and LED, obviously there can be overkill processors that will blink it just as well as a simple circuit but take way more power so it’s probably best to focus on what processors can have libraries that play multiple simultaneous sounds as a starting point and then look at their power requirements if there are multiple chips that are capable of playing the sounds as intended
@@GadgetReboot Thanks for the quick reply. Everything you mentioned made a lot of sense. I looked closer at your sketch and realized that the Teensy is not playing .wav files from an SD card, but converted .wavs using wav2sketch. So that effectively eliminates the need for an SD card, and allows you to store the sounds directly on the Teensy's flash memory. Really neat stuff! I wonder if you happen to know the overall size of the sketch in KB as it was uploaded to the Teensy? I'm trying to come up with a minimum RAM requirement for my project while choosing a microcontroller. I don't see myself needing 8 different sounds as you use in your project, but knowing how much ram your sketch required will help inform my decision as to what microcontrollers might be appropriate for my project. As I mentioned previously, I'm trying to keep my project compact physically, as well as keep the current draw to a minimum, so if I can get away with the 2MB of flash memory available on the smaller Teensy 4.0, I would probably prefer that over the larger Teensy 4.1, even though it has 8MB of flash.
I just compiled it and it shows Sketch uses 2159648 bytes (26%) of program storage space. Maximum is 8126464 bytes. Global variables use 45748 bytes (8%) of dynamic memory, leaving 478540 bytes for local variables. Maximum is 524288 bytes. So just over 2 MB of flash including all those samples and 45K of ram it looks like
@@GadgetReboot Awesome thanks for those numbers! Today I ordered the Teensy 4.0, along with their Audio Adapter Shield. Hopefully I can fit my program into the 2MB of flash on the Teensy 4.0, but if not, the Audio Shield has a nifty spot to solder on an additional flash memory module, which looks like could expand storage by another 128MB, which would be PLENTY of space for my use. Thanks again for the info. I'm considering creating a PCB on PCBway for some of the additional circuits I may need. If so, I'll be sure to use your affiliate link!
Hi my name is Eddie I’m a looping musician I use boss 505 loop station currently I’m using a foot switch to start loops and stop loops I’m trying to find a method to use a drum trigger to replace the foot switch and trigger the loop station with a drum pad
Looking at the graphic on this dual footswitch, which the site says is compatible, www.sweetwater.com/store/detail/FS6--boss-fs-6-dual-foot-switch It looks like the footswitches are just open/close contacts between tip or ring, and the sleeve, and so depending how that's designed, if the common sleeve is actually circuit ground, a simple transistor could be used to bring one of the other tip or ring inputs to ground, which can be driven by anything. Or worst case if it's some weird circuit that needs those pins shorted when pressed, a relay could be used to connect them from another circuit, or a JFET type transistor.
@@GadgetReboot that would be so cool I looked in my forest Memes getting started in electronics book and didn’t see a circuit that I could configure rate and that way or maybe I just didn’t know what to look for could you recommend a circuit for that application
This is not a great way to do Electronic drums. Maybe in the 80's. Considering you have a teensy 4 it's best to do all that stuff in software and use advanced filtering so you can do waveform analysis and do positional sensing and whatnot. There is a lot of information that you lose by not giving the teensy a proper AC signal.
If it's good enough for the 80's, it's good enough for the workbench! A great way to do something is the way that can be implemented sufficiently and quickly within the time budget and technical capabilities of the team, and especially for hobbyists, captures the interest and enthusiasm of those working on the project. But advanced signal processing certainly has its place. Although by presenting an AC signal, we would have to bias it up to allow full swing, and thereby lose some information by further compressing it to fit within full scale of the input. Do you have a link to a reference project that does better signal processing on a piezo to show what type of extra useful detail can be extracted? I'm always looking to improve projects. Thanks for watching!
An amateur making drums (better than 80's drum machines) and walking you through the process is in no way an "non-great way" to do things. Thanks for coming out :)
@@SimpleElectronics It's OK but not great. I'm referring to what he did with the signal coming in to the teensy. The reason you want to do what he did is if you are limited in the number of analog reads you can do. The teensy 4 can easily do one million ADC reads per second and you can do advanced filtering (bessel, chebyshev, etc) and a bunch of other stuff while it's waiting for the ADC reads to finish.
@@sabahoudini The hardware interface can be used with other devices beside Teensy so being widely compatible is good. Also getting it done in hardware with a few parts vs advanced filtering, which I don't know if there's an off the shelf Teensy solution I can use off hand, well beyond the scope. But if you have links to reference material I'd like to look at it. And there's still the need to condition the analog signal to fit within 0-3.3v and bias up the AC so the negative part isn't clipped.
@@GadgetReboot I'm using an op amp as a buffer and a voltage divider as a virtual ground to get a midpoint between 0-3.3V. The op amp is connected to ground and 3.3v. I'm getting super clean signals using the ADC library from the teensy forum sampling at 12 bits at 50kHz and then filtering to get very high resolution using floats for everything. You can find everything you need on the teensy forum. I suggest looking for the ADC library where you can set more advanced ADC settings.
