|
|
View previous topic :: View next topic |
Author |
Message |
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Thu Nov 20, 2008 7:07 pm Post subject: Amplifier for low sensitivity DIY mods | Easy & Cheap |
|
|
Hello everyone and welcome to the thread you are looking for if you want to build a cheap amplification circuit for your note dropping under sensitive DIY drum set.
After 1 month of adjusting and redesign of a DIY drum set I made following several tutorials on these and other rock band related forums, mostly MikeJl46's tutorial at www.edrums.info I became frustrated with how my sensitivity was so low and dropped hits. I turned to the forums for another wave of redesign, trials and errors to finally obtain no statisfying results. The solution left was to amp the signal the heads generated, using a circuit, like the J5 interface. But the price tag, the fact that I can be pretty cheap and my die hard DIY spirit pushed me to find my own solution. The result is in this tutorial.
WARNING
THIS DOES NOT IN ANYWAY FIX A DEAD PAD, NO DETECTION AT ALL MEANS A BROKEN PIEZO, BAD CONNECTIONS, ETC...
USE AT YOUR OWN RISK, I CANNOT BE HELD RESPONSIBLE IF YOU FRY YOU SET, OR EVEN YOUR XBOX THIS WAY!
** English is not my native language so pardon my language typos and formulation, I did my best **
Difficulty :
On a scale of 1 to 5, where 5 is "hard as hell" to achieve, this mod is (and was for me anyway) a 2. Considering you are a DIY guy at heart and held at some point in your life a soldering iron (some basic physics/electronics class at some point in your life can help, but not necessary which was my case before I started this project
** Cough ** you need a bit of perseverance though, snags slow you down ** cough **
Application & Compatibility :
- This mod works with any console. It affects only the drum set, and it "does it's job" before the rock band "does his" (further reading explains).
- You can also use this same circuit to turn a broken/unused/defective set into a cheap rock band drum brain with adjustments for any standard e-drum set, using jack connectors and normal pads[/b]
- I strongly suggest avoiding this mod unless you plan on modding the drum beforehand (pads, rack, etc...). This mod isn't a night's work and fits a lot better in bigger DIY project.
A word about piezos and Rock Band 1 controllers :
*** COMING SOON *** Go here in the mean time : http://rockband.scorehero.com/forum/viewtopic.php?t=15457&start=255
The circuit :
The following schema is for a single pad. Each pad on the set needs this circuit
Now looking at this, if you have little to know knowledge on electronics, might scare you. Just take some time and it gets very easy to understand.
From left to right, you see the piezo, one of it's two cables (doesn't matter which) is plugged into the "non inverting input" (the +) of the amplifier. The other cable on the piezo goes strait to the RB chip, but it is grounded. It's important you put a 47K resistor between both cables like in the schema. The opamp is the triangle in the middle, it calculates how much it should amplify by comparing the input (the +) with it's output (the tip of the triangle) which is fed back into it. The output now goes to the RB chip and "Y"ed, divided and fed into the "inverting input" (the -) and grounded. By divided I mean that the two resistors at the bottom are dividing the voltage. The opamp is supplied by a a direct current between 3 and15V. See next paragraph for a more detailed explanation of the more "complicated" aspect of this stuff
Whats and Hows :
Now up till now it can look a bit complicated, but I assure you, this is quite simple, just requires you to take the time to do it, and since you're a DIY guy at heart, i'm pretty sure you want to know more :
- GROUND : In a direct current circuit like this one, the ground is basically the negative ("-", 0V, colored black) of whatever powers your circuit. The ground on the schema is illustrated with a striped arrow pointing down.
- OPAMP : The hardest part in this schema is the triangle shape component in the middle. It's called an Operational Amplifier (op-amp/opamp). On the pics (**comming soon**) it's the "insect chip" :P. These things come in a variety of shapes, sizes and configurations. I use two LM358s. Each chip contains 2 opamps in one and you can find which leg is what with this schema : http://www.mcu.hk/GIF/LM358.gif. Keep reading and you'll see that it's easy to plug all this together (considering you know how to solder stuff :P).
- DIRECT CURRENT : Goes in one direction, used heavily in electronics, but not so much in other cases where it's mostly AC power. This means it has a current node (more than 0V, the "+") and ground node (0V, the "-")
- PCB : printed circuit board, experts get these done custom for their circuits, but I use pre printed boards, cheaper and I don't need fancy circuitry for this stuff.
Parts :
This is the complete list of parts you will need to make the circuit for 4 pads, multiply the amounts if you want more pads, but remember the controller only has 4 inputs
- 2xLM358 8 pin IC OPAMP
- 4x47K resistor
- 4 resistors or any kind (I recommend 5K) with a trimmer pot with equal or greater max resitance (I recommend 50K to 100K) OR 8 identical resistors of any amount (I highly unrecommende that one). See below for an explanation
- One 9V battery if you plan on not using the USB 5V.
- Optional connection heads (male/female) for simple plugging/unplugging of the piezos, can be anything you want. You can also just solder them directly on the circuit.
- 4 Piezo sensors/the mesh heads
- The rock band set with internal cables at hand ready to be plugged/soldered into the circuit
- PCB board of your choice for soldering it all together, although you can see what configuration I took for my circuit below
- Cables with the same gauge as the resistor cables to make connections on the PCB. You can find copper cable of that gauge inside most Ethernet Cables if you are extra cheap, just be careful when you splice them open.
How it works :
This is a very basic op-amp circuit. It's called a "non-inverting amplifier circuit". You can read more on this stuff here :
http://en.wikipedia.org/wiki/Opamps
And base diagram is here :
http://en.wikipedia.org/wiki/Opamps#Basic_non-inverting_amplifier_circuit
http://en.wikipedia.org/wiki/Operational_amplifier_applications#Non-inverting_amplifier
What this circuit basically does increase the signal that the piezo sensor generates. This signal is a tension variation (voltage). A simple formula helps know the gain of the circuit :
R1 = left bottom resistor
R2 = right bottom resistor
GAIN = 1 + (R2/R1)
What you need to understand is that the two bottom resistors are what decide how much the signal is amplified(gain).
Let's say, for example, you only need to double the signal of the piezo, just get two identical resistors of any resistance value and the signal will be doubled. Here is the math if you take two 10k resistors :
GAIN = 1 + (10/10) = 2
Now having a static configuration is not a good idea for many reasons, depending on your set and how you built the modded heads, sentivity adjustments might differ from head to head. The solution is a trimmer pot at the right position (R2, like in the diagramm). A trimmer pot is basicly an adjustable resistance. Now lets say you want a gain range from 1 to 5, you need a trimmer that has 4 times the maximum value of the left resistance. Here is the math again if you take a 10K resistance on the left, you then need a 40k trimmer pot on the right for a 1 to 5 gain range :
GAIN = 1 + (X/10) where X is the trimmers set resistance
If you set you trimmer at 30K :
GAIN = 1 + (30/10) = 4
The maths for this are really simple, just making sure it's crystal clear :P
For logical reasons you cannot go as low as 1 gain, which is at the same time what we want :P. It's also important you don't confuse the right and left trimmers (like i did :P) because division by zero is impossible... and that is BAD. I highly suggest a voltmeter/multimeter to test your output compared to a normal un-amped head. Depending on how you built your head, it may be extremely noisy (loose mesh head) and cause a "flood" of signals and the rock band set just starts ignoring the hits you do (rolling and successive hits might drop, even when hit EXTREMELY HARD). The head may also be extremely hard to trigger, the case usually is that your head is very tight and all the vibrations are dissipated and the piezo doesn't generate enough of a signal for the set to detect. My best solution (i tried almost everything :P) is a VERY TIGHT mesh head and amplify it a lot.
I'm not sure if all the sets are consistent in how the cabling to the piezos is, but I BELIEVE it's safe to hit and miss, the piezo can be plugged in both directions, doesn't change anything, but how the set is plugged to the circuit does make a difference :P. The circuit above is how I plugged it and it works...
The links I posted earlier have a lot of info on the tests that have been done so if you require more information, I suggest reading those or ask away on this thread
Suggested disposition for PCB :
I made the following to give a disposition if you choose the same PCB layout I have. This is for only 2 pads, since these opamps can only amp 2 piezos a piece.
** I CANNOT STATE THAT THIS IS THE BEST DISPOSITION, IT'S DONE BY A COMPLETE NOOB, BUT IT WORKS, SUGGESTIONS AND CONSTRUCTIVE CRITICISM IS WELCOME **
Just repeat the pattern as many times as you need. If you don't get it, it's 6 times the same picture, just with added layers of parts and complexity.
Pics of my circuit :
Costs (all in canadian $, sorry...):
Each :
- The LM358 Opamp | 1$ max, (0.76$ for me)
- Trimmer Pots with handles | 2.50$ (ouch, yah, most expensive part)
- Resistors | let's say, 0.10$ each, between 0.5 and 0.12$ though depending on resistance, etc.
- The PCB | 2 to 4$, depends on model, configuration, etc...
- The optional connectors | 1$ per kit (male, female, and metal innards) maximum
- Some cable | with 2$ bucks you'll have well enough, maybe double that if you are sloppy.
- The battery if you dont want to use the USB power | 1$ cheap stuff to good stuff 10$ for a 9V
The most expensive you can get with
- 4 pads
- Each pad has his trimmer
- 1 PCB
- Nice 9V battery
- 8 Connectors (4 for the pads into the circuit, and 4 to plug the RB stock cables into the circuit)
Totals to 37$, AND I AM VERY GENEROUS ON THE PRICES.
Remove the battery and it's immediatly 10$ off, screw stock cable connectors and that's 4$, bargain a bit and easily another 3$ gone.
The bang for your buck in that case is 20$. I keep connectors for the heads so i can unplug and adjust easily.
This is not very expensive IMO, compared to orderable solutions. Elbow grease is needed nonetheless :P
Additional info :
- I highly recommend if you are a solder noob like me to build a prototype for one or two pads (since one opamp IC has 2 sets of amps). Test it out on your comp with the "drum machine" app or with the freestyle mode in RB2. You should also visually compare signals with a multimeter (analog is better for visualization, needle jumps is better than LCD numbers) or oscillator to see the difference and how the trimmer is affecting the gain
- Trimmer pots are odd to the ignorant hobbyist like me, they usually have 3 nodes while you'd expect 2... Most trimmers you'll find have to be plugged with either "side" nodes with the "middle" node :P. Use a multimeter if you aren't sure. Plug it up on resistor mode, and start turning the well/handle/screw and see if the resistance varies. Take note that direction doesnt affect resistors so you can plug it as you see fit. Resistors are like dams, they just "resist" :P All my trimmers have one of the "side" nodes connected to nothing.
Credits :
Here's the original thread that started the subject on these forums :
http://rockband.scorehero.com/forum/viewtopic.php?t=17735
- First off is MikeJl46 and his DIY drum tutorials. He has great stuff at edrums.info and anyone who want's to make solid heads for their drums should have a serious look at the site
- These forums, for the info I found on them prior to posting this stuff.
- Electronics-Lab.com, great site for help on electronics and projects, my thread for this circuit http://www.electronics-lab.com/forum/index.php?topic=16014.0
- Wikipedia, cuz it's awesome
Further developments :
- If there are any theory experts on these forums, a bit more explanations would not harm this thread :P I am a computer programmer with a brand new hobby in electronics. So this is still very new to me.
- Might look into a pedal addition...
- FAQ if people ask questions
Last edited by Malchom on Tue Jul 14, 2009 5:19 am; edited 20 times in total |
|
Back to top |
|
|
SixStringedBass
Joined: 14 Jul 2007 Posts: 181
|
Posted: Thu Nov 20, 2008 8:46 pm Post subject: |
|
|
As for the trimmer, I believe that you need to switch it with the other resistor that is inline with it in your picture. Also, I don't know if it would work, but check out the LM124 or even the LM324. I believe that it could work for what it is that is needed. |
|
Back to top |
|
|
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Fri Nov 21, 2008 5:39 am Post subject: |
|
|
SixStringed you are right that diagramm is wrong, i'll fix it :P
I'll be adding a circuit representation soon also :P
Dunno if this is eligible for sticky? |
|
Back to top |
|
|
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Fri Nov 21, 2008 8:14 am Post subject: |
|
|
I updated the first post a bit more, now that I've tested and switched the bottom resistors. More coming in the future :P |
|
Back to top |
|
|
nosmorking
Joined: 23 Nov 2008 Posts: 1
|
Posted: Sun Nov 23, 2008 2:02 am Post subject: PS3? |
|
|
I see that this is set up for Xbox. Will it work on a PS3 also? |
|
Back to top |
|
|
SixStringedBass
Joined: 14 Jul 2007 Posts: 181
|
Posted: Sun Nov 23, 2008 3:18 am Post subject: Re: PS3? |
|
|
nosmorking wrote: | I see that this is set up for Xbox. Will it work on a PS3 also? |
I don't see why not. This only affects the amplitude of the piezo, it has nothing to do with the way the controller interacts with the system. |
|
Back to top |
|
|
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Sun Nov 23, 2008 4:30 am Post subject: |
|
|
Yup, this mod works with any set console wise.
It amplifies the signal before it reaches the chip in the set.
I also updated a bit more the thread and changed the title of the thread to remind that this is still not over :P |
|
Back to top |
|
|
SixStringedBass
Joined: 14 Jul 2007 Posts: 181
|
Posted: Sun Nov 23, 2008 9:16 pm Post subject: |
|
|
Somebody correct me if I am wrong, but doesn't the LM324 work with voltages as low as 3v? If that is the case, then couldn't you simply use the LM324 with the supplied 5v from the usb cable in the controller? |
|
Back to top |
|
|
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Mon Nov 24, 2008 2:45 am Post subject: |
|
|
That's what I have been wondering, but is there enough power and will it affect the stock circuit? |
|
Back to top |
|
|
akwok
Joined: 30 Oct 2007 Posts: 199
|
Posted: Mon Nov 24, 2008 2:55 am Post subject: |
|
|
Malchom wrote: | That's what I have been wondering, but is there enough power and will it affect the stock circuit? |
Well a lot of us run MSA-P boxes straight out of the USB 5v, and it works fine! |
|
Back to top |
|
|
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Wed Nov 26, 2008 1:15 am Post subject: |
|
|
I am buying the new parts for my circuit tomorrow or Thursday. I will be trying to use the USB power as well and will be posting and updating the results
So far everything is working top shape. Although I have tested various tensions on the mesh head and so far best results are with a very tight head with a lot of amplification.
Loose heads create a lot more vibration than wanted and the op-amp gets really excited and the signal can get a bit high. A high signal seems to drop hits, like if the chip has a window in which a variation can be analyzed. I will be making further tests on this.
Also, only a gain of 2 isn't enough, not nearly enough actually with a tight head. I changed the 100k resistor with a 10K one and detection is nearly spot on with a full gain (10 in this case). I will be trying with a smaller resistor which will me give me a bit more gain and see what happens
Also, with a tight head, the signal is a lot weaker, because the tightness of it all makes it vibrate less. Unlike metal, plastic or rubber, the mesh heads and the wood seem to dissipate vibration quite easily. This is great considering it lowers the chance of double hits but also a hassle since the signal is much weaker (I can barely see something moving on my meter).
I tried one of my heads on a friend's set and it detects hits much better than mine (even withough an amp... it's a QM), sadly buying a new set destroys the point of this mod of mine :P I WILL NOT GIVE UP! |
|
Back to top |
|
|
akwok
Joined: 30 Oct 2007 Posts: 199
|
Posted: Wed Nov 26, 2008 1:32 am Post subject: |
|
|
Alright, sweet, can't wait! My DIY mesh heads right now are dropping notes and aren't anywhere near sensitive enough |
|
Back to top |
|
|
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Mon Dec 01, 2008 1:18 pm Post subject: |
|
|
I am almost done with the circuit. Had a few things in the way slowing down the hole process
At the same time I am done I will prbly post an update on my findings on making good triggers.
Also! LM324 DO NOT WORK... Dunno why but they just don't do work... |
|
Back to top |
|
|
Malchom
Joined: 02 Nov 2008 Posts: 156 Location: Montréal, Québec, CANADA
|
Posted: Wed Dec 03, 2008 3:30 am Post subject: |
|
|
Alright, I made the near final update for this thread!
Only thing left is a FAQ, but I can only make that one if people ask questions.
I will be adding pics of the final product.
If you try this out and have questions ask away, or else I'll let it wither and die away slowly on these forums _________________
|
|
Back to top |
|
|
SixStringedBass
Joined: 14 Jul 2007 Posts: 181
|
Posted: Wed Dec 03, 2008 8:57 pm Post subject: |
|
|
One thing that I noticed in your op is that if you want a range of 1-5, you actually want a potentiometer of 4x your R1 value due to the fact that amplification is defined as 1 + r2/r1. So having a 5x r1 value will give you a 6x range. |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Copyright © 2006-2024 ScoreHero, LLC
|
Powered by phpBB
|