Announcement

Collapse
No announcement yet.

Odd hotspot on A2E snare

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

  • Odd hotspot on A2E snare

    I have a 14" A2E snare with Pintech Reaction mesh head and Ufodrums trigger . The trigger is mounted to the lugs at about 6 o'clock. I don't have their crossbar. The rim piezo is mounted to the shell at about 5:30 (so right next to the lugs holding the head trigger.)

    The head is pretty dang tight.

    While playing I would notice some random spikes in volume. What I figured out was that at about 12 o'clock there is a hotspot that sounds like I'm hitting right on top of the head trigger. The hotspot area is actually fairly large just smaller than say a coffee cup.

    How is this possible?

    If I play in the center or over towards 3 o'clock or 9 o'clock I get the normal volume but if I play up around 12 o'clock or 6 o'clock I get the hotspot spikes.

    Thoughts? Ideas?

    TD-25 module. PS turned off.

    Thanks.
    Last edited by billgtx; 12-02-18, 07:11 AM.
    ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

  • #2
    Hi Bill, is the head tension even all around? If you have a "node" of tension across one or two lugs it can do strange things across the head, particularly with larger drums. 13" is the largest I go with DIYs, partly for portability and partly for this reason. I also religiously tune my mesh heads for the best, even performance all around, particularly with edge triggers.

    Make sure the rim is the same distance off the bearing edge. Also, is the foam cone perhaps too high? (A rule of thumb is ~3 mm or 1/8".)
    ATV aDrums & aD5, Pearl Mimic Pro & DIY, Agean R-series Silent Cymbals, Roland Handsonic HPD-20.

    Comment


    • #3
      Hey JP. I'll double check all of that. I generally go by feel when I turn the lugs but I'll try to pay more attention. I'm pretty sure the cone height is good. But I'll check again. Probably remove the head and start over.
      ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

      Comment


      • #4
        Took the head off. Checked the cone height. Looks good. Checked connections. Reseated the head and used the tuning trick of tapping near each lug. Even mesh heads produce a little resonance. Tightened each lug half a turn (star pattern). Did this 3 or 4 times. Checked by tapping after each round.

        Then turned on the module and tested. Still have a hotspot at 12 o'clock. Fiddled with tension on the 3 rods in the area. No change to the hotspot.

        I even took the head off again and rotated it. Reseated and retightened. Still have the hotspot.

        Next chance I get I'm going to move the trigger to a different lug to see if the hotspot rotates with it.

        Any other ideas?
        ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

        Comment


        • #5
          Everyone doing DIY pads should have an oscilloscope. Period. Without one you're just wandering around in the fog. And there's nobody who can help you because you can't drill down to the actual problem.
          Digital storage scopes have become pretty cheap these days, and old analog ones are often given away for almost free. If you still don't have one in your lab, feed the audio in a DAW/Audio Editor in your PC/Mac and check the waveform there. You may also check with a scope VST plugin for better visibility.
          Looking at the actual waveform (or a series thereof) makes it more easy to specify whether it's a reflection pattern, mechanical dampening issue, resonance byproduct or whatever.
          MarkDrum YES e-kit highly modified (DIY hall-sensor based hihat, low-volume trigger cymbals, 16" DIY kick, 12" DIY snare + tom 3), Triggera 10" splash
          Gibraltar 9607NL-DP Legless Hi Hat, Intruder Double Pedal
          Shure SE215 in-ears w. CustomArt silicone tips

          Comment


          • #6
            All right. Supposing I want to play scientist. I have Win10 and Audacity. Will that work for this experiment?

            What am I looking for?
            ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

            Comment


            • #7
              Connected my snare to the laptop through the mic port and collected some samples from Audacity. What you see in the screen shot is even strokes from the center of the snare gradually moving to 12 o'clock. The trigger is mounted to the lugs at the 6 o'clock position, so the strikes are moving away from the trigger to the opposite side of the drum.

              Thoughts?
              Attached Files
              ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

              Comment


              • #8
                Interesting problem you have there....

                How is the isolation between the shell and the piezo. It almost looks like you pick up the shell vibration.

                Can you show us a picture of the trigger construction, with the head removed?
                Brain: mega drum. 5 toms: DIY mesh head, side-mounted DIY triggers. Snare: 14" 682 head, DIY crossbar trigger. 2xDIY beaterless BD pedal. .Cymbals: Crash: 2x 16" brass: 2 zone. Ride: 20" brass: 2 zone. Hi-Hat: 14" 1 zone DIY Control pedal + Pearl H900 stand. + drum rack:

                Comment


                • #9
                  It's natural that the piezo signal becomes more sharp/spiky towards the rim (which is what Roland use to detect position, it's the 'slew rate'), and since the head is stiffer along the rim, waves travel faster there. Therefore it doesn't matter much where along the rim you put the piezo.
                  But looking at the center-position waveform, I think there's way too much ringing and an overly long decay. For a trigger circuit, the best signal would be a sharp spike along the entire surface. A trigger algorithm has an easy job when it just needs to look for a spike within a given time window (or upon zero-crossing and counting up time from there). If the spikes (aka slew rate) aren't high enough, it needs to check over a longer period and use better methods like taking the difference of two envelopes (which results in a delta signal), but that causes additional latency, so most modules omit it and do it the 'dumb' way. Therefore it's important to make it easier for the module on the mechanical side.
                  I suppose you're still using a foam cone? I made good experience with cylindrical form for rim placement, but it's important to find cell foam that has a closed surface on top then, since open form acts as a lowpass and 'swallows' light hits. A cone is more broad-band or even in this respect, but on the rim its shape is counter-productive when it comes to ringing: a rim trigger works best when the rim is dampened (slightly). Plus, cones produce less output since the contact area is way smaller. And the ratio of area to voltage isn't linear, it's exponential.

                  What you could try is putting some foam ring (maybe 1cm wide or so) along the rim under the mesh, which would kill the decay immediately. That area shouldn't be too large, otherwise it would hinder the ways from getting to the piezo directly, but a small circle area should already be an improvement. What you're looking for is 'direct' waves hitting the sensor, but right now it's all kinds of reflections coming in there, especially when you hit the center or intermediate positions (causing interference patterns).

                  Oh, and you need to make sure you're stopping shell vibrations at the piezo. That should reside on a cushion. Problem is that the cushion under its 'butt' acts as a brake for harder hits, therefore it affects the overall dynamic behaviour in a negative way. And it can produce ripples when it bounces back on heavy hits that make the head dive in more (supposedly in the center). One has to find a sweet spot for the cushion here, between minimum shell crosstalk, minimum bounce, and maximum dynamics.

                  Since this task can be tedious, you might want to observe it in realtime, using an oscilloscope plugin. Audacity can host VST & AU plugins. Check out this page, it gives a good overview on oscilloscope tools for your platform: https://www.kvraudio.com/forum/viewtopic.php?t=471835
                  Last edited by sascha; 12-05-18, 04:21 AM.
                  MarkDrum YES e-kit highly modified (DIY hall-sensor based hihat, low-volume trigger cymbals, 16" DIY kick, 12" DIY snare + tom 3), Triggera 10" splash
                  Gibraltar 9607NL-DP Legless Hi Hat, Intruder Double Pedal
                  Shure SE215 in-ears w. CustomArt silicone tips

                  Comment


                  • #10
                    Ok. That's a lot to digest so let me start by saying 2 things.

                    I didn't show the picture but the waveform at the 3 o'clock and 9 o'clock positions resemble the center waveform more than the 12 o'clock waveform. So its more than just being close to the rim.

                    Secondly I'm using ufodrums trigger assembly. Not of my own making. It is the upgraded conical cone not the octagonal cone. The rim piezo is attached to the rim with the provided double sided tape.

                    I'll try to get some more pictures this afternoon after work.

                    I have an extra Triggera InTrigg I believe so I might put that in and see what it does.
                    ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

                    Comment


                    • #11
                      If 3 & 9 o'clock look more like the center one it's even worse. IMO the decay of the waveform should be way shorter. It looks noisy, reverberant. Try making measurements before & after some dampening, at the exact same location and - as much as possible - the same tension.
                      MarkDrum YES e-kit highly modified (DIY hall-sensor based hihat, low-volume trigger cymbals, 16" DIY kick, 12" DIY snare + tom 3), Triggera 10" splash
                      Gibraltar 9607NL-DP Legless Hi Hat, Intruder Double Pedal
                      Shure SE215 in-ears w. CustomArt silicone tips

                      Comment


                      • #12
                        More data. I've attached 5 screenshots. Four of them are individual hits blown up (trigger is UfoDrums conical side trigger). I then switched out the UfoDrums trigger for the Triggera InTrigg which uses a cylinder style foam. The last attachment is several hits in the center of snare and a few at the 12 o'clock position from the InTrigg. Looks like the same phenomena.

                        i then played the snare on my module. There is definitely a hot spot centered about 2 inches from the 12 o'clock position. No where else except right over the foam gives this hot spot.

                        Thoughts?

                        Edit - forgot to mention. I haven't had time to look into dampening. I did tighten the head a little more - beyond my comfort zone.
                        Attached Files
                        Last edited by billgtx; 12-05-18, 02:28 PM.
                        ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

                        Comment


                        • #13
                          Make some measurements after dampening along the rim. Could be interesting.
                          I have no idea how long the tails are currently. Can you give some info on that? It's hard to see it from the waveforms since they're zoomed out so much. I mean something like RT60 (decrease in level by 60dB) as the decay time. Ideally should only be a few milliseconds.

                          If there's still no improvement after dampening and changing triggers and positions it *must* be the head. BTW, there's no use in making it overtight. You just make the medium stiffer, so that waves can travel faster. That can improve tracking by making the attack sharper or more distinct and predictable, but at the same time diffuse-field reflections from straying echoes along the surface create more noise that the tracking has to cope with. So stay in your comfort zone, otherwise it would be a waste of time and effort.
                          MarkDrum YES e-kit highly modified (DIY hall-sensor based hihat, low-volume trigger cymbals, 16" DIY kick, 12" DIY snare + tom 3), Triggera 10" splash
                          Gibraltar 9607NL-DP Legless Hi Hat, Intruder Double Pedal
                          Shure SE215 in-ears w. CustomArt silicone tips

                          Comment


                          • #14
                            Originally posted by sascha View Post
                            Make some measurements after dampening along the rim. Could be interesting.
                            I'll look into that. May take some time to figure out what I need to get and get to it.

                            Originally posted by sascha View Post
                            I have no idea how long the tails are currently. Can you give some info on that? It's hard to see it from the waveforms since they're zoomed out so much. I mean something like RT60 (decrease in level by 60dB) as the decay time. Ideally should only be a few milliseconds.
                            I'm not sure what you're asking. When I look at the Snare 12 O'Clock picture I see the strike occur at about 4.935 seconds and you can see the waveform die out around 5.5 seconds. Or are you looking for something different?

                            Originally posted by sascha View Post
                            If there's still no improvement after dampening and changing triggers and positions it *must* be the head.
                            I'm using a single ply Pintech Reaction head. I'll figure out how to center mount the trigger and remeasure. I had another thought. Would it help to take my Roland PDX-100 and do the same test?

                            Thanks for walking through this with me. In the end I'm probably going to end up with a center (or near center) mounted trigger and/or using a log function curve on the trigger settings to minimize the hotspot but I'm willing to keep looking at this.


                            ATV aD5, TD-25KV cymbals, Mapex Horizon Blue Sparkle shells A2E w/Triggera InTriggs.Yamaha DXR15, MG10. Senn 280HD.

                            Comment


                            • #15
                              Wow, more than half a second of decay? if it really needs that time to decrease by 60dB, that's huuuge! I'm not sure what's 'usual' but I would try shooting for like 10% of that time. Definitely try dampening that guy. And yes, try another pad or mesh, just to have more data to play with.
                              MarkDrum YES e-kit highly modified (DIY hall-sensor based hihat, low-volume trigger cymbals, 16" DIY kick, 12" DIY snare + tom 3), Triggera 10" splash
                              Gibraltar 9607NL-DP Legless Hi Hat, Intruder Double Pedal
                              Shure SE215 in-ears w. CustomArt silicone tips

                              Comment

                              Working...
                              X