Maker Pro
Maker Pro

1V audio burst every OTHER pushbutton

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
Chris, what if it isn't a mic? I have a piano pickup that works like an electric guitar pickup, it had no moving parts, it isn't an acoustic device... it works by picking up the metal strings moving within its magnetic field. It is made my Helpinstill if you want to look it up... it also ignores phantom power just like an SM58 mix. think this would solve the problem?
 

Laplace

Apr 4, 2010
1,252
Joined
Apr 4, 2010
Messages
1,252
Is there a way to only trigger the circuit when the pedal is moving above a certain speed near the end of the release?
Such a thing is possible. It would require two lift sensors (probably precision optical) near the bottom of the drop stroke. But then the question is, since the overdrive pulse is supposed to be audibly imperceptible, then what difference would it make?
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
Sorry I didn't quite understand what you were asking just now, the overdrive pulse?
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Chris, what if it isn't a mic? I have a piano pickup that works like an electric guitar pickup, it had no moving parts, it isn't an acoustic device... it works by picking up the metal strings moving within its magnetic field. It is made my Helpinstill if you want to look it up... it also ignores phantom power just like an SM58 mix. think this would solve the problem?
I'm not certain I can answer this accurately. I'm not that familiar with the hookups that musicians typically use. Though I would think that it would be desirable to have all audio inputs independent of each other. Doing so eliminates loading effects that parallel wired devices would certainly experience. It also makes adjusting audio input levels independent without affecting one another.

I believe most dynamic mics have a Zo in the range of 600Ω to about 1KΩ. I have no idea what the impedance of the pickup you mention is.

Chris
 

Laplace

Apr 4, 2010
1,252
Joined
Apr 4, 2010
Messages
1,252
What if you fed the mic signal into a pre-amp, then to a passive mixer to add the 10kHz burst (adjusting the pre-amp gain to offset the mixer loss), then to the compressor and low-pass filter? Seems like it would be a simple task to design the passive mixer as part of the burst device, maybe even include the pre-amp as well. But I am unfamiliar with the techniques for handling Phantom Power in audio amplifiers.
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
right, i see. the thing is, I'm having to do this for ten grand pianos, which already have three inputs each on my mixer (the bass channel, a left and a right). Adding another input per piano would bump me up to 40 channels, and I also have to mic up a string quartet, several vocalists... I'm going to run out of channels quickly. That's why i'd like to figure this out by utilizing the already existing bass channel where the thumping is occurring.

Question: is there simple summing circuitry which can combine the outputs of the piano mic/pickup and the 10kHz burst while keeping them isolated from one another? Much like two channels on a mixer board?
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
Without some kind of velocity sensor, the 10kHz burst will trigger a quick compression on the mixer which will mute the piano signal every time the dampers come down on the strings. This is probably fine, but there are some times when piano players will very slowly and lightly let up on the pedal and use the dampers to slowly mute the strings... which would not cause a thump but would trigger the short compression and might cause an awkward break in the sound. So I thought maybe if there was a simple way to only trigger the circuit when the pedal is being released at higher than certain speeds, it would assure that the signal is only muted when there would actually be a thump... and of course it needs to be about 30dB louder at the mic pre than the loudest parts of the piano songs, so that the compressor threshold can be set higher than the piano so as not to be triggered by it, but be set lower than the 10kHz burst so that it is triggered by it.
 

Laplace

Apr 4, 2010
1,252
Joined
Apr 4, 2010
Messages
1,252
What I'm thinking of for a velocity sensor would be two photo-transistors illuminated by two LEDs, spaced about 3/16" apart vertically. A shade attached to the lift pipe will descend and block the light for one and then both photo-transistors. The bottom photo-transistor will be aligned with the bottom of the lift pipe travel range. Therefore the time it takes for the pipe to travel down from the top to the bottom photo-transistors indicates the drop speed of the damper just before it hits the strings.

For the timing circuit the top photo-transistor will trigger a monostable pulse (duration TBD), and if the descending shade reaches the bottom photo-transistor before the pulse terminates, then the descent speed is faster than the set limit, and that condition will trigger the 10kHz burst. For best accuracy the photo-transistors must be illuminated through an optical slit (0.01" wide or less).

While I have a few ideas (like insuring the optical slits remain clean), it would be best to leave the design and construction of the optical sensor to others.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
What I'm thinking of for a velocity sensor would be two photo-transistors illuminated by two LEDs, <<<snip>>>

Brilliant! I love the concept! Devin's project is now growing a brain.

Chris
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
Wow that sounds like an awesome idea! Is there the potential for this optical sensor to be falsely triggered in concert situations where there is a lot of stage lighting and effects? What kind of chassis are we thinking? And just curious, but is there a reliable way to do this with magnets?
 

Laplace

Apr 4, 2010
1,252
Joined
Apr 4, 2010
Messages
1,252
False triggering by stage lighting - Depends on how you build the sensor. I imagined mounting the LED & photo-transistor on opposite sides of a 1/4" brass tube and cutting the optical slit with a razor saw. A polished stainless steel rod would fit inside the brass tube as the plunger attached to the lift pipe. So the only light would come from the LED with no chance for stray light to interfere.

Using magnets instead of LEDs - The problem would be getting the magnetic field focused for accurate position sensing. Maybe you could do that by machining soft iron rod down to a chisel shape to focus the field for the magnet and Hall sensor. Then an accurate position could be determined when the chisel ends are directly opposite each other (with a gap of a few thousandths of an inch). That setup would need two Hall sensors but just one magnet.
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
This really is a great help to me so far, I appreciate everyone's input! Would somebody be willing to scribble out a quick diagram of the circuit and the parts needed?

Also, is it possible to achieve the same triggering effect with only one LED and two phototransistors? Can the circuit be made to only trigger when the LED passes in front of the top and then the bottom phototransistor, and only within a certain timeframe (which can be selected maybe by use of a pot or slider, from 0-500ms)? And would it matter whether the LED in nominal position rest directly in front of the bottom phototransistor or just past it?
 

Laplace

Apr 4, 2010
1,252
Joined
Apr 4, 2010
Messages
1,252
From your questions I would surmise that you have decided not to employ optical slits with a plunger for light cutoff to the sensor. Since I have no familiarity with pianos of any kind or their damper mechanisms, it is difficult to imagine what can be achieved otherwise.

Does the damper descend by gravity alone or is it under the force of spring tension? What is the range of motion of the lift pipe, whose position & velocity is what I assumed was to be measured. What is the maximum velocity of the lift pipe just prior to the damper hitting the strings?

One of the reasons for accurately measuring velocity at the bottom of the range is that the circuit would not need enabling by the top sensor because, from your description, the operator (player?) might lift the damper high yet slow its descent prior to bottoming.
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
I think everyone here has made it clear that optical is the best solution, so that's the direction I'll go for sure. The pedal is spring loaded, but the velocity is completely determined by the player throughout the cycle (a human isn't capable of lifting their foot faster than the spring will counter-react). The total cycle range up and down varies slightly by piano, but is approximately 1/2" ...and the pipe may also shift axis a bit during the cycle and not move perfectly up and down. Every piano is different, and I'll just have to figure out some mounting hardware which gives me options and play room.

So at this point I'm wondering if there's a generous person out there who'd be willing to do a quick sketch of the circuit this far, with the single LED and two phototransistor a, the timer pot for selecting the timeframe from 0-.5s within which the bottom phototrans must be triggered after the top one is triggered, the delay of the 10kHz burst from 0-.5s once it is triggered, the volume control from 0-1.4V RMS on the balanced output, the necessary circuitry for sine-ifying the burst and limiting it to 5ms in length, the ability to use the 48V Phantom Power from the audio mixer to power the circuit, and then some kind of summing circuitry to be able to combine the 10kHz burst with the signal from the piano pickup or mic without either one affecting/damaging the other.

If the summing part is too much trouble, leave it out and I'll just use two inputs on my mixer, one for the pickup/mic and one for the 10kHz burst...

Once again, the big idea here is that the piano player will depress the pedal, causing the LED on the pipe to move upwards about 1/2" and past both phototransistors, and then when the pedal is released, the LED will first move past the top phototransistor, and then the second a short time later. This time span can be arbitrarily determined, and only if the movement from the top to bottom phototransistor happens within it is the burst triggered. This burst will be used to trigger software on an audio mixer which temporarily mutes the piano pickup/mic because the dampers make a loud thump when they come crashing down on all 88 strings during most normal to fast pedal releases.

Thanks in advance, and thank you all so far for the great input and suggestions!
 
Last edited:

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
Rather than use two photo-transistors and a moving LED, which would eventually lead to fractured wiring, have you considered using two static photo-interrupters and a moving light-blocker?
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
That would work as well! The circuitry would be designed the same way either way, right?
 

Laplace

Apr 4, 2010
1,252
Joined
Apr 4, 2010
Messages
1,252
Here is an example of what I meant by using optical slits. This is a 9/32" OD brass tube with a 1/4" wood dowel. The slits are separated by 3/8" and measure 0.35mm wide. Opposite each slit is an opening for LED illumination. Still left to do is solder smaller brass tubes to hold the photo-transistors & LEDs. Looking to get a photo-transistor in a TO-18 can. The 9/32" brass tube would be mounted on the PCB, and the wood dowel replaced with a brass or aluminum rod to be attached to the damper lift pipe. At this point it's not clear whether activation should occur when the sliding rod blocks or unblocks the optical slits.
IMG_2.jpg

Image1.jpg
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
are those pre-made, or do I have to make them myself?
 

devin

Jul 28, 2015
21
Joined
Jul 28, 2015
Messages
21
So I'd mount the dowel or blocker on the piano bar which would move up upon depression, exposing both phototransistors, and then as the pedal is released, the dowel/blocker will first block the top LED and then the bottom LED a short time later. The circuit needs to ignore other combinations as well, such as the pedal only being depressed enough that the bottom phototransistor is exposed but released back to nominal position never having exposed the top phototransistor. So if I use either this pipe with the wooden dowel, or the static photo-interrupters, what kind of circuit diagram would give me the 10kHz burst only when this condition is fulfilled within a selectable amount of time from 0-.5s?
 
Top