Announcement

Collapse
No announcement yet.

Concept Thread for Roland Optical Hi-Hat Controller

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Originally posted by John F View Post
    The problem with this circuit is that the 3.2 volts you show coming from the Roland module is not a "stiff" 3.2 volts. If you hook up the circuit you've shown to the Roland module, the voltage will drop to about half of what you think.

    I took some measurements on my TD3 and here's what I found. It's very likely that this will apply to other Roland modules as well because most likely the HH interface hardware is reused and any differences among the various models are in software algorithms. No, I don't work for Roland, but I've designed electronic products for long enough to know that this is generally the case .

    Voltage out of TD3 (foot pedal not plugged in) = 3.26 V
    Voltage at foot pedal (pedal plugged in/full up) = 2.55 V
    Voltage at foot pedal (pedal plugged in/full down) = 5 millivolts
    Pedal resistance (full up) = 20.2 kohms
    Pedal resistance (full down) = 7.4 ohms

    From these measurements the TD3 looks like a 3.26 volt source with a 5.6 kohm series resistor, so if you hook up your circuit as shown and assuming that your photoresistor value is large compared to the 4.7k resistor, the voltage at the top of the photoresistor will be set by the voltage divider made up of the internal 5.6k resistor and the 4.7k resistor plus the base-emitter drop of the transistor, and will be somewhere around 1/2 the original voltage.
    So this circuit won't work because of the voltage divider. I understand that. Is there an acceptable value to replace the 4.7K resistor with that would work? I tried the circuit with a 10K, 4.7K and 1K and they all worked. The problem still seem to me that the voltage when the hat is completely closed or there abouts would not be enough to bias the transistor.
    Last edited by jman 31; 05-20-09, 06:49 PM.
    sigpic JerEd Systems, LED drum triggers

    Comment


    • #17
      Originally posted by Hellfire View Post
      I always thought that guy might be using FSR's and you just confirmed it for me.

      So, I guess the FSR method works pretty easily then (if he is using an FSR) Can you post a picture of the inside of what you are talking about? (please use the topic below if you are going to post further about FSRs or softpots)

      jman 31, I want to apologize for hijacking your topic. It was not my intension to do this. I ask that if you want to talk about the FSR or SoftPot concept please do so in this topic

      http://vdrums.com/forum/showthread.p...322#post393322
      No problem man. I don't have a big ego. I am just looking for another way to accomplish a task with less chance of it breaking. That's why this is a concept thread. I like the optical for a quick way to convert an acoustic hat, but it doesn't have to be the end all be all.
      sigpic JerEd Systems, LED drum triggers

      Comment


      • #18
        Jman

        Another contactless type controller is a Magnopot.

        http://www.spectrasymbol.com/typo3/s...gnetopots.html

        Pretty cool, but a bit expensive...

        http://www.sparkfun.com/commerce/pro...oducts_id=9072

        Comment


        • #19
          Originally posted by macca2004 View Post
          Jman

          Another contactless type controller is a Magnopot.

          http://www.spectrasymbol.com/typo3/s...gnetopots.html

          Pretty cool, but a bit expensive...

          http://www.sparkfun.com/commerce/pro...oducts_id=9072
          Man, I really like that idea. When I get some time I will sit down and give it a thorough looking into. I think this would solve the problem of having an active source and needing power to an led.
          sigpic JerEd Systems, LED drum triggers

          Comment


          • #20
            Originally posted by Hellfire View Post
            I always thought that guy might be using FSR's and you just confirmed it for me.

            So, I guess the FSR method works pretty easily then (if he is using an FSR) Can you post a picture of the inside of what you are talking about? (please use the topic below if you are going to post further about FSRs or softpots)

            jman 31, I want to apologize for hijacking your topic. It was not my intension to do this. I ask that if you want to talk about the FSR or SoftPot concept please do so in this topic

            http://vdrums.com/forum/showthread.p...322#post393322
            I'll crack one open this weekend. I have mine sandwiched between my converted 3z Hats, but since I have only a TD-3 I have no idea what it's potential is. I did the 3z conversion looking toward future module upgrade this summer. Looks very simple even to me though. I was thinking of converting my HartWare(switch) footcontroller to emulate a FD-8 style pedal and then convert a stand for my 3zHH. I have pics of my coversion process(ASSEMBLY) I will post them.
            8 piece DIY Acrylic, 2x2Box DrumIt5, Gen16 4xDCP, DIY Acrylic&Gen16 Conversions, Sleishman Twin-QuadSteele hybrid, Gibraltar&DrumFrame rack, DW9502LB, Midi Knights Pro Lighting
            http://www.airbrushartists.org/DreamscapeAirbrushRealm

            Comment


            • #21
              Originally posted by jman 31 View Post
              So this circuit won't work because of the voltage divider. I understand that. Is there an acceptable value to replace the 4.7K resistor with that would work? I tried the circuit with a 10K, 4.7K and 1K and they all worked. The problem still seem to me that the voltage when the hat is completely closed or there abouts would not be enough to bias the transistor.
              Yes, it will work - it's just that the voltage divider kinda screws up the voltage to the circuit. And if (or when) the photoresistor value goes low enough, you're right, there won't be enough voltage to turn on the transistor that controls the LED.

              Comment


              • #22
                Jeremy,

                I have a couple ideas that I'm toying around with but haven't done much with yet. I don't think the photo resistor is a viable approach because I don't think you can get the resistance low enough. My FD8 measures about 7 ohms at full down and 20k at full up. Ideally, the FD8 should provide zero ohms at full down. I haven't looked in great detail but I don't think you can get the resistance to vary over that great a range.

                I'm going out on a limb here and I'll try to explain what I believe is how the Roland module works with the HH. This is based on my measurements of the TD3 and FD8, plus some experience in designing circuits that do essentially what the module is doing in the HH interface circuit.

                There seems to be the opinion out there that the Roland modules read the resistance. Not true - you can't directly measure resistance, all you can directly measure is voltage. (You could apply a constant current source and measure the voltage then calculate the resistance but that's not what they're doing, nor is there any need to do so). The Roland module uses the FSR in the FD8 to vary the voltage at the input of an (internal to the module) analog to digital converter (ADC). The ADC is read at regular intervals. The voltage reading at any given time is the position of the foot pedal. The change in voltage from one reading to the next is the velocity of the moving pedal. From this, the software can compute whatever it needs to (hats part way open, volume of "chick", etc.) To do this they (TD3) have a 3.3 volt source with a 5.6k resistor in series hooked to the 20k FSR. The 5.6k resistor protects the 3.3V source from being shorted out by the user. It also makes the voltage at the 20k FSR = 2.55V for full up. This is really convenient (although not entirely necessary) if the ADC is 8 bit, because now 1 bit will be 10 millivolts (8 bit ADCs are very common, i.e., cheap).

                So, theoretically, all you need to do is supply 0V (for full down) to 2.55V (for full up) to the input of the TD3 and you're good to go. The problem is that inside the module there's that 5.6k resistor and 3.3V source. That means that you need to drive the module input from a very low impedance so that the voltage you supply isn't affected by the internal 3.3V. I've run a couple simulations of circuits that will do this, but what I don't have a feel for is what kind of voltage change you're getting on your optical HH controller from your previous thread. Do you have a schematic of the circuit and a measurement of full up and down output voltages? I think I might be able to interface this to a Roland module.

                Comment


                • #23
                  Thanks for the feedback John. I have also been researching this issue and came to the same conclusion about the resistance not being the factor, but the voltage. I was starting to look at the phototransistor again instead of the photoresistor simply for pure speed.

                  If you are asking about the schematic for the controller that we are using for the megadrum then here it is.



                  I have been giving the exact thing you said some thought also, about not affecting the 3.3v or having it affect my circuit. I just don't have a clue how to implement it. I would love to see you ideas!

                  Jeremy
                  sigpic JerEd Systems, LED drum triggers

                  Comment


                  • #24
                    Originally posted by John F View Post
                    Jeremy,

                    I have a couple ideas that I'm toying around with but haven't done much with yet. I don't think the photo resistor is a viable approach because I don't think you can get the resistance low enough. My FD8 measures about 7 ohms at full down and 20k at full up. Ideally, the FD8 should provide zero ohms at full down. I haven't looked in great detail but I don't think you can get the resistance to vary over that great a range.

                    I'm going out on a limb here and I'll try to explain what I believe is how the Roland module works with the HH. This is based on my measurements of the TD3 and FD8, plus some experience in designing circuits that do essentially what the module is doing in the HH interface circuit.

                    There seems to be the opinion out there that the Roland modules read the resistance. Not true - you can't directly measure resistance, all you can directly measure is voltage. (You could apply a constant current source and measure the voltage then calculate the resistance but that's not what they're doing, nor is there any need to do so). The Roland module uses the FSR in the FD8 to vary the voltage at the input of an (internal to the module) analog to digital converter (ADC). The ADC is read at regular intervals. The voltage reading at any given time is the position of the foot pedal. The change in voltage from one reading to the next is the velocity of the moving pedal. From this, the software can compute whatever it needs to (hats part way open, volume of "chick", etc.) To do this they (TD3) have a 3.3 volt source with a 5.6k resistor in series hooked to the 20k FSR. The 5.6k resistor protects the 3.3V source from being shorted out by the user. It also makes the voltage at the 20k FSR = 2.55V for full up. This is really convenient (although not entirely necessary) if the ADC is 8 bit, because now 1 bit will be 10 millivolts (8 bit ADCs are very common, i.e., cheap).

                    So, theoretically, all you need to do is supply 0V (for full down) to 2.55V (for full up) to the input of the TD3 and you're good to go. The problem is that inside the module there's that 5.6k resistor and 3.3V source. That means that you need to drive the module input from a very low impedance so that the voltage you supply isn't affected by the internal 3.3V. I've run a couple simulations of circuits that will do this, but what I don't have a feel for is what kind of voltage change you're getting on your optical HH controller from your previous thread. Do you have a schematic of the circuit and a measurement of full up and down output voltages? I think I might be able to interface this to a Roland module.
                    Hey John,

                    Did you ever give this any more thought? I am really interested in pursuing this. I think that a Yamaha Module may work with something relatively close also after looking a schematic for a Yamaha pedal.
                    Attached Files
                    sigpic JerEd Systems, LED drum triggers

                    Comment


                    • #25
                      Hey guys, I am back on this kick again. I haven't had much time to mess with this idea so i am going to try to give it some more thought.

                      I have a question concerning a Roland FD-8 pedal. Do they have a mono jack? And is the input on a Roland Module a mono input?

                      Jman
                      sigpic JerEd Systems, LED drum triggers

                      Comment


                      • #26
                        Originally posted by jman 31 View Post
                        Hey guys, I am back on this kick again. I haven't had much time to mess with this idea so i am going to try to give it some more thought.

                        I have a question concerning a Roland FD-8 pedal. Do they have a mono jack? And is the input on a Roland Module a mono input?

                        Jman
                        I was wondering what happened to this thread. Glad to see you're back at it.

                        That is a good question. Although I do not use a FD-8, I do use a TD-8 module. I just switched out the TRS cable with a mono cable and it worked exactly the same as far as I could tell.
                        sigpic

                        Comment


                        • #27
                          Originally posted by cheapthrill View Post
                          I was wondering what happened to this thread. Glad to see you're back at it.

                          That is a good question. Although I do not use a FD-8, I do use a TD-8 module. I just switched out the TRS cable with a mono cable and it worked exactly the same as far as I could tell.
                          Thanks Cheapthrill, I was assuming that a mono plug would work fine since people are using a pot with only two wires to control it.
                          sigpic JerEd Systems, LED drum triggers

                          Comment


                          • #28
                            It's mono, if you plug a mono cable into the module with the tip and sleeve wires laid bare and cross them the hi hat chokes, add a variable slider pot in between the 2 and you have the closed to open stages.
                            WEBSITE - http://www.diamondelectronicdrums.com/

                            YOUTUBE CHANNEL - http://www.youtube.com/channel/UCbVB...?feature=guide

                            FACEBOOK me at ... https://www.facebook.com/photo.php?f...83235555050736

                            :eek: ...
                            Showcase 1 - http://vdrums.com/forum/album.php?albumid=253
                            Showcase 2 - http://vdrums.com/forum/album.php?albumid=354

                            Comment


                            • #29
                              Responding to John F's eighth post down - revealing piece of work! For an optical approach a CDS cell is waaaay to slow and as mentioned elsewhere does not present a low enough resistance value. The best chance for an optical approach is to use a photodiode in the photoconductive mode. From the numbers John F is showing, the current into the module input at full down pedal is close to 570 microamps - within the capabilities of a photodioe in the photoconductive mode. racer52 has fooled with an interesting implementation but he indicated it did not work with his Roland module. I don't know sufficient details of his circuit but I do feel that the photoconductive mode will work. Whatever circuit is used should have the effect of presenting to the module an input voltage ranging from about 0 to +2.55 VDC.

                              Comment

                              Working...
                              X