Help building a single piezo drum pad

Hi all, I'm trying to check around to see exactly what materials people have used to create their own DIY drum pad projects.
I am using a ceramic piezo sensor and the entire disc is about 1" diameter. I will attach a product link if that helps, just let me know.

This is a follow up on a post I made under project guidance so I apologize for the double post but I can't seem to get a reply.

From what I can tell, foam is advisable underneath whatever the piezo sensor is attached to in order to control extra vibrations from drum hits (ideally I want to be able to use drumsticks on this).
I've also seen rubber recommended but I'm not sure exactly what I'm looking for when I go to the craft store.
I'd also like to ask if you guys think metal is required because I have seen a couple of examples where the sensor itself was attached to the center of a metal disc.

This is a prototype with the goal being drumsticks hitting whatever the drum head is made of, and sending voltage to my Teensy 4.0 to transform into MIDI velocity values.

Does anybody have advice on the materials to use? Thank you so much for reading and for any info anyone can provide!!
 
It's difficult to predict how good or bad foam, rubber and metal discs will work together. Since that's all mechanical interaction, it's best to experiment with different materials and find out what works best for you.

Paul
 
Just to show you: this afternoon I built a simple drum pad using a piezo transducer, a 110mm PVC cap and an EPDM rubber sheet on top.

drumpad.jpg

A Teensy LC samples the transducer signal and sends a MIDI note to Ableton Live.
Here is the code I used for a first test:
Code:
#include <MIDI.h>

void setup() {
}

void loop() {
  int piezoADC = analogRead(A2);
  if (piezoADC > 150) {
    usbMIDI.sendNoteOn(14, 127, 10); // note D0(14), D1(26)
    delay(100);
  }
  delay(5);
  while (usbMIDI.read()) {  // read & ignore incoming messages
  }
}
The pad works reliable with drumsticks or hands. It's not velocity sensitive yet...more coding to be done.
Some pratical notes:
1. glued the piezo transducer to the PVC cap using cyanoacrylate.
2. added 2 Schottky diodes (1N5817) to protect the Teensy input from a too high (>3V3) or too low (<0V) transducer signal.
3. placed a 1Mohm resistor over the transducer wires.

Paul
 
Heya, Paul! This was exactly the kind of information I was hoping for. I just need an idea of a direction to go in and I'm okay if it doesn't work out perfectly. My brain kind of got overloaded with all the different approaches and materials people were throwing in to make their own drum heads even if they were only making one versus a whole kit.
To that end I have to say thank you so so much for providing the example that you have. I hope it doesn't seem like I'm looking to have my hand held...I know I'll be making plenty of mistakes as I go so the learning is guaranteed, lol.

If I may, I was hoping to ask you a few questions about the setup that you've described there. Originally, I saw your initial reply to me which linked to this thread: https://forum.pjrc.com/threads/48308-test-code-for-piezo

In that thread you described what sounds like a thorough and cautious way to build a circuit and keep it under control so that there's negligible risk of damage to the Teensy board. I pictured the parts listed such as the 470 ohm resistor and the 10K ohm resistor, as well as diodes that were a specific number you mentioned (1N4148). Belatedly I noticed that thread was from 2017, and I think I would have a much easier time going along with the example you posted now.

I'm guessing the parts that I got previously will not apply for the example you posted the other day - I have no 1Mohm resistors but I can get some, as I'm guessing the gap in what protection they would provide is considerable. Also the type of diode you recommend in the example from the other day is still a Schottky type but I suspect I need to order those separately (the 1N5817) because again I assume there's a reason for the different component recommendations with this newer example. Also looking at their data sheets there was a pretty big difference in reverse voltage capability between what I got for the 2017 example and for the type your recommending here.

Sorry for being a long-winded, I genuinely apologize but I want to try and get this as correct as I can because I know unexpected things will pop up anyway.

Finally a couple of miscellaneous questions, can you tell me the diameter of the PVC cap that you got, and do you find it to be a comfortable size?
Is the placement of the sensor intended to be in a certain quadrant so that other sensors will be added for your particular project? I've always assumed I would just need to center mine and I'm only trying for one drum trigger to begin with.
In the last question is for the EPDM rubber sheet would you recommend any particular thickness?
I can't think you enough for the information you provided and for your patients with me. I really am trying to pull my weight as far as researching and not just depending on others for answers but I also truly appreciate being able to reach out and discuss this with someone so knowledgeable.
Thank you kindly for your time Paul. Cheers!
 
You're welcome.
In the thread you are referring to, it was Paul Stoffregen, founder of PJRC, who replied several times - not me :)

With respect to the Schottky diodes: I felt the 1N5817 to be a safer choice do its lower Vforward than the 1N4148 - so less risk to kill a Teensy (pin).
The PVC cap I used was actually 125mm [5"]. You can find it in a hardware store at the plumbing department.
I indeed preferred to place the piezo transducer in the center of the cap but that wasn't possible due to internal ribbing...so the sensitivity over the pad area is not consistent now.
Is 125mm comfortable? Don't know - I can imagine a real drummer finds this a bit small. But my purpose was just to try something.
The EPDM rubber I used is 1.1mm thick. Had a few pieces left over after roofing a shed. Just thought it felt nice for drumsticks.

Success with experimenting!
Paul
 
Yes its best practice to use external Schottky diodes to protect internal protection diodes - the internal ones are Si pn-junction so will still carry current if in parallel with an external Si pn-junction. Schottky's being metal-silicon junctions have a much lower forward voltage and take all the current. You'll see commercial designs are littered with dual Schottky's like BAT54S's and diode/zener array chips like USB6B1 to protect CMOS logic lines.
 
Thank you so much, both of you for your in-depth replies. I cannot wait to get this going. Right now I think I've got everything lined out except, I don't yet have a good way to get a small quantity of EPDM. I can only find rolls which are humongous and kind of expensive. Is there any alternative material that you feel might have similar qualities, bearing in mind that this is still my first attempt and just a prototype overall?
Thank you both so much for your time, Paul and MarkT.
 
The way high end drums are made is with a 27mm piezo mounted on a 22mm dual sided foam tape circle. Thickness of the foam about 2mm, glued with the dual sided tape on a firm surface with the brass side of the piezo on the dual sided tape. On top of the piezo (on the ceramic part) is a poron foam cone. This is mounted inside a drum with a mesh head on it, The foam cone is touching the mesh head from the bottom and presses on it with about 1.5mm sticking above the bearing edge.

The poron cone is usually 6 layers tall and every poron sheet is 6mm tall so 36mm cone. Poron is used because it keeps it's properties even after prolonged compression. (regular foam deforms into the compressed shape and stays like that.

Anyhow, if you do now wish to make a mesh head drum it's still good to know that the best way to mount a piezo to anything is with a dual sided foam tape about the size of the ceramic part and about 2mm thick.This is so the piezo can bend freely and give a good signal.
 
Frank! Thanks for your reply. That makes a lot of sense. I actually didn't consider how important it might be that the sensor can actually move/bend (if I'm understanding you're correctly.)
I really appreciate the in-depth breakdown of a professional approach to this project and some of the things that you've mentioned, I've seen suggested here and there especially the cone apparatuses. What I'm hoping to try and do right now is use very basic materials, achieve a result that doesn't seem hopeless, and still use some fairly makeshift/cheap materials at least for the first stage.
I will start looking for the dual-sided foam tape. That's an invaluable tip for me right now. I've been paralyzed with silly indecision about what materials to scoop up and just put the darn thing together.

The biggest question area I guess I have right now is about sandwiching the materials.
Right now my plan is to use the PVC cap from Paul's idea in the same thread, although I have not found what feels like an exact replacement for EPDM because they sell it by the roll here so 30 to 50 to 100 bucks USD. I don't know anyone who will rip me off a sheet, lol. But I have made over a dozen phone calls trying to find such (like from some contractor who wouldn't mind just tearing off a sheet or two).
In lieu of being able to find the EPDM,
I've seen this suggested a few places: mouse pads.
That may be what I go with in place of rubber for now unless anyone has any better advice.
And I am all ears for any thoughts you guys or anyone else has! Thank you so much for the feedback on this.
My finger has been poised on the trigger to just buy some of these materials and I know the initial product won't be perfect (and it will probably be laughably imperfect! Which is good) but I have a serious OCD over-mind that kicks in and tells me I have to wait until I've got some sure bet as to how to build the drum in a way that makes the sensor and the signal and everything all work in a fairly optimal way together. But this is my first project. So I do realize how illogical that is and I'm trying to short circuit that OCD bs. I know I won't ever learn if I don't just take some sort of plunge here.
Again, I can't tell you how much I appreciate you guys weighing in on this and if you have any further thoughts, even if it's just something like hey stop talking about it and just go get some materials whether the right or wrong, any and all communication about it is appreciated!
Cheers and I wouldn't be getting as close as I am to doing this without you guys on this forum.
 
I was going to follow up with one more question... The leads from the piezo sensor are (probably notoriously) tiny and malleable. What do you guys think is a best or at least reasonable practice for securing them in...
the little DuPont black terminals, and also
all the header slots on a breadboard?
I can deal with typical colloquially referred to Dupont stuff but the cables that are coming off of the sensor are honestly too small for my fingers to really manipulate effectively. xD
How do you guys deal with this in hooking them into your microcontroller or bread board?
 
The way high end drums are made is with a 27mm piezo mounted on a 22mm dual sided foam tape circle. Thickness of the foam about 2mm, glued with the dual sided tape on a firm surface with the brass side of the piezo on the dual sided tape. On top of the piezo (on the ceramic part) is a poron foam cone. This is mounted inside a drum with a mesh head on it, The foam cone is touching the mesh head from the bottom and presses on it with about 1.5mm sticking above the bearing edge.
Thanks Frank, for that usable info. I never dove into this matter but this enlightens a lot.
Found some more pictures and tutorials:

MBconepiezos.png

FS7OI2QKBP31NAA.png

What I showed in message #5 is merely a rude piezo-based knock-sensor...

Paul
 
This thread is blossoming out into something way more informative than I could have ever dreamt of!
If you guys will bear with me here, I'm still sorting out what AWG numbers correspond to 'DuPont' or 2.54mm pitch. I guess it's because a lot of variables can go into that such as stranded or solid wire, etc, but I'm trying to nail down a cheap but sufficient crimping and stripping tool so that these wires become manageable.
There is one local product that stops at AWG22 with some caveats as far as insulation, etc, and I'm strongly debating going to buy it and just see if it works. Before I do that, I'd like to link it here and see if you guys have been through this before or if I need to just go on Amazon and get something that goes up to gage 24 or 26...

Option 1: cheap: https://www.harborfreight.com/electrical-crimper-and-stripper-57595.html

Option 2: bit pricier but also ratcheting (if that is an important factor). https://www.harborfreight.com/elect...pers/heavy-duty-ratcheting-crimper-58325.html

As far as listed gauges, both of these stop at 22, but I'm curious if it's still possible for something like one of these to work. If anybody can offer any advice I'll probably name my firstborn after you because I'm getting really close to just getting going on this.

You guys have been so incredibly kind and helpful. Thank you more than I can say.
 
Those crimpers are for this sort of connector. Is that what you are after?
If you are after something to create DUPONT connectors then you should be thinking of something like this.
An easier and cheaper solution would be to buy some of these, then cut one, or two, in half and solder to your piezo wire.
I would suggest putting some heat shrink on the wires so that you can insulate them after soldering them.
 
Thank you! That makes a lot of sense. And you're right, I'm not really wanting to mess with all the other types of connectors. I actually have a box that has the male and female crimp rolls, as well as a lot of the plastic black pin enclosures to house them in. And I think I have either that exact kit of wires from Elegoo, or basically the perfect equivalent of it.
Heat shrink makes sense because extending a wire by soldering sounds wonderful, but I wasn't exactly sure what tidied it all back together.

As far as the Glarks crimping tool, is that when you would personally endorse or are you just kind of showing an example? From what I can tell it's as good a brand as any and the price point is still good. But I've also read about 5 million reviews of crimping and stripping tools and people's experiences seem to be all over the place. I'm glad you pointed me towards that because that definitely seems to be the tool for this job. And, I know I'll probably want to keep messing around with this for years to come after this initial project.

So would you recommend the Glarks or was that just an example and should I try to research something more specific to my needs? I wouldn't really know how to define those needs at this moment so if that particular model seems decent to you then I'm going to just take a chance on it because any of the models I get are probably going to be somewhat of a chance. I am very comforted by the fact that it goes up to gauge 18 and now I don't have to worry about bolting out and buying something totally mismatched for what I'm trying to do, although I will admit to totally being ready to jump in my car and book it just to get it today. Lol. But I guess learning and growing with the stuff has to kind of take its own time.

I think my only other question is if I need a wire stripper? I haven't measured how much uninsulated I have to work with on the piezo cables but as you have one yourself, you can probably imagine it's not much. Also, unless nippers are indispensable for some reason, I have a pair of wire cutters from years back when I was making wire jewelry and I think they'll be fine.

Thank you so much for putting up with my stream of consciousness questions. I haven't gone at something with this much fervor in a long time and it's truly interesting how many different possibilities and variabilities there are. At the same time, there's the desire to zero in on a 'best practice' for every decision about this project and I have to let that go. Experimenting and failing are things that I want to reclaim as fun because whenever I started making music digitally, I was using Fast Tracker 2 for DOS and wasn't the least bit worried about doing anything other than goofy experimentation which, luckily in inadvertently, led to learning.

Anyway, thank you so much everyone who has been responding because this is a dream of mine that I'm finally going to be able to make happen after several years of just wanting to.
You guys are amazing and I'm so glad that I happened to get into this community and got a Teensy board.

I'm truly grateful for you guys' Patience and thoroughness with sharing your experiences and knowledge. Whenever I felt like I would never figure any of the stuff out, just for this project, it was kind of nightmarish feeling but now it honestly feels like it's a dream that's going to come true in the next few days or, week, you know, some frame of time. Lol.
 
Thank you for all three of these links, BriComp! Ironically and amusingly, my friend from the UK is familiar with EPDM in his line of work and he told me that they sell it in lesser quantities there and so your timing on that is impeccable, lol. Apparently UK denizens are just way more chill about their EPDM supplies than anyone here in the States...haha. Regarding the drum heads I will definitely check those out. I could see one of those working into 'phase 2' after I've built my initial proof of concept out of sticks and mud. I won't begin to try and enumerate the hours I've spent googling frenetically over all sorts of aspects of what's involved in this project. That's been going on a month or more, and that times I've paralyzed myself thinking the 'right answer' just wasn't going to become apparent. Anyway, really do appreciate your feedback and I'm going to save these links.
 
As far as the Glarks crimping tool, is that when you would personally endorse or are you just kind of showing an example?
The GLARKS was just by way ofv example.
There is a lot of Chinese sellers selling similar or the same crimpers.
I usually try and look at the recommendations (or otherwise) for any purchase off Amazon, and just remember caveat emptor.
 
Understood. I've already made a couple of purchases that I later realized were not what I needed at all, but I think they're things I'll end up using. And actually the best example of that is a kit with the male and female crimp connectors, although I think I want to try and get fairly adept at the method of joining pre-made Dupont wires that you mentioned, as well. Heat shrink is awesomely cheap.

To that end, and again thank you so much for your patience with all these questions, would the method you outlined necessitate wire strippers to expose enough conductor to join together? Maybe as I get more practice at soldering It won't seem as necessary but it feels like it's about a third of an inch of copper poking out of each of those wires from the piezo.

I'm guessing if one were to really get going with this process and needing a lot of wire, the spools of colored wire would eventually become cost-effective? But I definitely appreciate you scaling your recommendations for this first project I'm doing because it's got a pretty small scope, haha.

Thank you kindly once again for your time and all of the information. Can't reiterate that enough. :)
 
Wire Strippers, I would recommend something like this.

As for joining wires.
Twist together the two wires to be soldered.

You will now have the wire ends to be soldered sticking at right angles to the general run of the wires.

Solder together.

Bend the soldered wires so they become parallel with the run of the resultant cable.

Now move the heat-shrink over the joined wire joint and shrink.

You will not end up with the nicest looking joint but it will be a strong succesfull soldered joint.
 
Okay, just to make sure I'm following you correctly here, twist the two cables together at roughly a 90° angle protruding from the normal flow of the wire, solder them together, and then bend the resulting mass back in line with the axis of the cable? Then heat shrink and voila?
 
Back
Top