How the Drums work (Mostly)

**K9Intimidation** · Joined: 05 Dec 2007 Posts: 1

I have been lurking here for a while and have not seen any particularly accurate information on how the drum kit works so I did a little investigating on my own.

Before reading any further understand that any “suggestions” or possible hacks that I make are merely rhetorical and I take no responsibility for the carnage that will likely ensue if you use any of this info to modify the Rock Band drum kit.

First let’s look at the foot pedal as how that works is well known. The pedal does use a simple magnetic switch in the base and a fairly strong and small magnet at the very tip of the orange part of the pedal. Digi-Key has many variations of these switches and the magnets to go with them. It is actually a cleaver design to get around the longevity issues of a mechanical switch people were stomping on. If only the plastic parts were a well though out. The only problem I have with the design is the switch closes about 1 cm before the rubber bumpers hit the bottom. This puts a delay in the physical feedback of the beat to where beat is actually registered. I definitely notice it when playing heel down.

The drum pads are plastic with rubberized surface for the hitting area. The rubberized coating is quite thin and looks to be the same process as is used on many electronics for improved grip. The pads are resting of foam and are held in place by rubber rails. On the underside on the pad a thin lollypop shaped piezoelectric sensor that is held firmly against the plastic by a spring and, on my set, an orange retainer.

I find the drums sturdy enough when playing the game but when moving it around it feels flimsy. I am also a little disappointed in the surface coating they used. A thin rubberized layer is not enough to absorb much of the sound of beating the crap out of a piece of plastic with wooden sticks. I have done the foam/felt mod and it helps but I don’t think the foam is going to last long as I can permanently depress it with my finger. I am looking for a cheap source of colored neoprene 1.5-2.0 mm thick to try.

Now that you have been completely bored by reading the same things as in 50 other posts lets look at the piezo sensor and what it outputs. If you don’t know what a piezo is Wikipedia has a good article on piezoelectricity here:
http://en.wikipedia.org/wiki/Piezoelectricity
In brief the pizeo is a crystal that deforms when compressed and generate electricity. When you hit the pad with the stick it causes the pad to vibrate and that energy, in part, gets transferred to the pizeo.

The first image in a hit on the pad at near the threshold of what it will register as a hit. Peak to peak it is only about 4 volts and has mostly recovered in about 60 ms. Also it is resonating at roughly 150 Hz.

The second image of a hard hit with over 40 volts peak to peak and is still vibrating more that a soft hit 60 ms later. The hard hit also initially resonates at a slightly higher frequency than a soft hit at about 200 Hz. One interesting note is the 3 volts of DC offset after the hit. This should not matter if the sensor they are using is AC coupled or is just looking at rate of change but it is interesting.

The third image is a soft hit and a hard hit on the same screen to compare the difference.

The fourth is a close up of the initial waveform of a hard hit.

The fifth image is a capture of some single handed hits over 2 seconds and the last image is of 2 handed hits over 4 seconds.

What does it all mean?

Well, we know that the Rock Band pads use the same technology as digital drum pads. It is possible that if a real drum pads produces output in a similar voltage range and at similar frequencies it is likely that they would be compatible. (Please refer to my second paragraph) Unfortunately I do not have access to any drum pads or a desire to spend $100 or more on one to find out.

I can also speculate that the Duct Tape method of fixing the fast fill problem works because it is increasing the damping of the pad and shorting the time between a hit and when the piezo stop oscillating. This also leads me to believe that they are not looking directly at the waveform through an analogue to digital converter but rather have a discrete circuit that converts the oscillating waveform into a level to drive an input to a micro controller. This discrete circuit could be getting saturated by fast had hits and never have a chance to open up again thus missing all subsequent hits. This would be easy to confirm but I am unwilling to disassemble my new drums to that extent.

If the drums do use a discrete circuit to convert the waveform into a level signal it might be sensitive to the frequency that it is oscillating at and this might cause a problem in getting a real drum pad to work.

When looking for waveforms of real drum pads I did find some interesting sites dedicated to DIY drum pads. Particular of interest was the Remo/Regal practice pad conversion.

Computer Game Drum Kit $80
4-8” Mylar practice pads $80
4 thin metal sheets $20
Your friends shaking their heads at how you could turn something as cool as drumming into a nerd fest – Priceless.

http://www.edrum.info/pads.html
http://www.toontrack.com/forum/tt.aspx?forumid=52

**BriGuy** · Joined: 04 Mar 2006 Posts: 1894 Location: Boston

Very interesting research. This means that fast parts should be hit softer for a better signal.
The other thing that might be a helpful mod is adding more padding underneath the pedal to bridge that 1cm gap if you want a more tactile feel to your stomp.
_________________

**zombie1942** · Joined: 09 Apr 2007 Posts: 241

My brain just exploded and I crapped my pants both at the same time...

I have to also believe that its seems that the majority of the pads affected are the red and green because they are out on the "arms" of the set with nothing under them to dampen the vibration of fast rolls... The tape method worked for me so good that its almost scary.

**OpenYourEyes** · Posted: Sun Dec 09, 2007 1:45 am Post subject:

Wow thats some interesting stuff....

_________________

blingdomepiece wrote:

Eastwinn wrote:

As a kid I wasted my time cheating in SimCity or the Sims, and now that I'm discovering how much fun it is to not cheat, I'm also discovering how stupidly hard it is

Set all the tax rates to 9.

**evilgoat** · Joined: 09 Apr 2007 Posts: 722 Location: Toms River, NJ

This is both excellent and amazing. This is why I treat the kit as a controller and not as a simulation. Just like vocals, there are tricks that can help you beat the hard parts. Extensive research, thanks for sharing!
_________________

XBL ragnartheawful PSN Ragnar_the_Awful

**kckondor** · Joined: 03 Aug 2007 Posts: 41

I don't know if anybody has seen this, but on one of the loading screens, it says something about playing the fast parts softer for better accuracy, which the test seems to confirm.

Very interesting.

**evilgoat** · Joined: 09 Apr 2007 Posts: 722 Location: Toms River, NJ

**disquette** · Joined: 12 Dec 2007 Posts: 82

Hey K9, what software and hardware did you use to do your measurements? I'd be interested in doing this sort of measuring in conjunction with the various penny fixes to see how the waves differ between the configurations.

TIA!