MIDI concertina III: the load cell version

[Richard Mellish]

17 hours ago, RAc said:

 

Agreed. The root idea is to approximate the behavior or a real concertina just enough to use the MIDI as a usable training instrument (noise reduction being the key factor here). My dev work will come to an abrupt stop exactly at a point where I can use the gizmo for the daily drill so I can impress the girls with my killer skills without exposing my beloved to my routine practicing.

 

Hope that makes sense! 😉

If you feel you really need not only to practise pushing the buttons but also to get the feel of bellows movement rather than just applying push and pull forces, then you will indeed need something along the lines you envisage. That is probably easier to build than real bellows with an air pressure sensor as adopted by some.

 

Coming the other way, from donkey's years of playing real concertinas to building a MIDI one, I find the minimal movement in and out not a problem at all and even a benefit as I can keep pulling or pushing in the same direction for as long as I like; although in practice, simulating an Anglo, I need to reverse direction fairly often anyway.

[Richard Mellish]

One further thought. I already pointed out that you need to provide an appropriate amount of resistance to movement. The speed of movement should be (at least roughly) proportional to the force, as with real bellows. Slide bars with friction would not achieve that. Telescoping tubes with a narrow gap for air to flow in and out might make a reasonable approximation. You could perhaps use an off-the-shelf gas spring (I've seen a YouTube video about how those work but I couldn't find it in a quick search.)

 

Preferably the speed for a given force should also depend on how many buttons you press at once, but that would be more difficult to simulate.

[caj]

On 11/3/2024 at 10:57 AM, RAc said:

unfortunately, the data sheet (HC-SR04-datasheet-version-2.pdf) states that the cycle period should not be below 50ms. It is probably still sufficient but may not yield a fine enough granularity for reliable velocity sampling. I will probably give it a try anyways.

 

 

I just tried one out in the lab, with a Nano and facing the sensor toward a sheet of phenolic (circuit board.)  I noticed that I can read as rapidly as 20ms; if I go down below that, the reader can give me spurious readings, which may be due to echoes from the previous sound.

 

I'm not sure how reliable that error is; if it just causes an occasional outlier value one might write code that just discards low return values.   Beyond that, the readings were quite stable, and would probably make a reliable bellows reading.

[RAc]

On 11/4/2024 at 6:12 PM, caj said:

 

I just tried one out in the lab, with a Nano and facing the sensor toward a sheet of phenolic (circuit board.)  I noticed that I can read as rapidly as 20ms; if I go down below that, the reader can give me spurious readings, which may be due to echoes from the previous sound.

 

I'm not sure how reliable that error is; if it just causes an occasional outlier value one might write code that just discards low return values.   Beyond that, the readings were quite stable, and would probably make a reliable bellows reading.

 

I have now ordered a few of the HC-SR04 modules along with VL53L0X and VL6180 infra red transceivers (I have a feeling that ultrasonic may not be the best of choices given that we have dogs). Will publish data and results once they come in.
 

[caj]

15 hours ago, RAc said:

 

I have now ordered a few of the HC-SR04 modules along with VL53L0X and VL6180 infra red transceivers (I have a feeling that ultrasonic may not be the best of choices given that we have dogs). Will publish data and results once they come in.
 

 

Wow, that's a great idea.  That should fix any problems that ultrasonic sensors may have with reflections off the bellows, and it looks like you'll have the freedom to poll them faster.

 

Indeed, the bellows should work to your advantage by blocking external IR.

[RAc]

I have now had a chance to experiment with the VL6180x ToF sensor. It appears to work like a charm, covers exactly the range needed for my bellows, reads reliably and fast enough. I will receive the last mounting components in a few days' time frame and will present the final assembly then. For the mechanical part I simply built a gussetless bellows which has the additional advantage of a concertina like look and feel.

 

The only issue then is the software. For my experiments, all I do is sample the current time stamp and sensor reading in every loop traversal, then compute the deltas for both between the current and last reading, then map the quotient to the permitted volume range of 1-127 and use that as the global velocity byte for the midi transmission. Lots of room for improvement and fine tuning, but probabaly I will leave it at that.

 

For the time being, thanks for all the input and feedback, stay tuned!

[Don Taylor]

23 hours ago, RAc said:

For the mechanical part I simply built a gussetless bellows which has the additional advantage of a concertina like look and feel.

 

RAc:  I am interested in knowing more about your 'gussetless bellows'.  Pictures?  Maybe a link to a maker site on how to make them.

 

I have a VL53L0X in my parts bin and might try your idea myself.

[RAc]

1 hour ago, Don Taylor said:

RAc:  I am interested in knowing more about your 'gussetless bellows'.  Pictures?  Maybe a link to a maker site on how to make them.

 

I have a VL53L0X in my parts bin and might try your idea myself.

yes, give me a few days, I am about to finish the Gizmo anyways and will then make a docu bash including a video.