Jump to content

Midi Hayden Duet? Assessing Demand


Recommended Posts

My minimum spacing is from button center to center in any direction, so it looks like this mechanism would work without modifications for the Hayden duet layout. Here's the schematic for the oscillator:

 

InductorOscillator.PDF

 

As for tuning component values, the 1M feedback resistor should stay the same. For the inductor, 150 turns is about as many as I could fit onto that form, but if you go down to 100 turns, the frequency increases. I'm using equal value capacitors, but there's nothing intrinsic to the circuit that requires the capacitors be equal in value. I didn't do much experimenting with the capacitor values and the value of the 200 Ohm resistor; once I saw a nice change in frequency and a stable oscillation, I stopped playing.

 

If you want to try the approach of sampling all of the oscillators with a fast processor, you will want to make sure that the oscillators are slow enough that you sample at least twice per cycle, or at least 3 MHz for the values shown. I had to go from 100 turns to 150 turns because I was cutting it close. However, you might also want to send the signals into the frequency counter input of a microprocessor, if your micro has them. It probably won't have 32 of them, so you will need to multiplex the inputs, in which case you want to run the oscillator on the _fast_ side (fewer windings, smaller capacitance value) so that you can get a nice high count (and therefore high resolution) in as short a time as possible before you move on to the next oscillator. I'm beginning to think that that might have been a better design approach than the sampling approach, but my design is locked into a nice expensive PCB and I'll keep pressing forward with it.

 

In a month or so, once I've debugged the design a bit more, I'll post everything I've got on my lab wiki. (I don't like to post untested stuff there.). I'll post a message here when I've done that.

 

-JIm

 

Link to comment
Share on other sites

  • Replies 159
  • Created
  • Last Reply

Top Posters In This Topic

So let's talk airflow. The big advantage of instrumenting an existing concertina action is that the airflow ought to feel about right. I had hoped to be able to roughly duplicate the airflow change with my plunger idea, but there just isn't enough space between the coil form and the plunger to approximate a real concertina. I noticed that some posters mentioned making or buying a pneumatic servo valve. Has anyone found anything good?

Link to comment
Share on other sites

Jim:

 

If it has to feel like an existing bellows then maybe you only need a single air hole with a pad on each side that can be activated with a relay whenever a key is pressed on that side. Basically, simulate the way that a pad is raised/lowered whenever a button is pressed/released. Maybe this is what you meant by referring to a pneumatic servo valve?

 

OTOH, I am not sure that it is really necessary to faithfully simulate the feel of real bellows as a means of controlling volume dynamics, especially on an EC or a duet.

 

What are you planning to use to sense the pressure inside the bellows? There are some interesting products out there, but they can be quite expensive.

 

Don.

Link to comment
Share on other sites

I'm planning on using the Freescale MP3V5004DP:

 

http://www.digikey.com/product-search/en?x=0&y=0&lang=en&site=us&keywords=MP3V5004DP

 

Not the cheapest part out there, but not the most expensive. I needed a part with 2 ports to sense the differential pressure, and a range of less than +/-1PSI. Also, it needed to work well at 3.3V, which is my system voltage. I've tested it statically, but I won't know how it performs for real until I can put everything together.

 

I hadn't considered how low pressure the inside of a bellows is, until I did the math and made some measurements -- less than 1/2 PSI differential pressure, and that's pushing hard.

 

I guess a simple solenoid to raise/lower a pad would make the most sense. Next time, perhaps.

 

-Jim

Link to comment
Share on other sites

  • 3 weeks later...

Making progress, despite a messy New England winter keeping me away from my lab. All of the chips are up and running, and now I have to make a decision that I've been kicking down the road for far too long. I've got enough DSP power on-board to do some reasonable 4-voice synthesis, and 4GB of memory to store samples. But now I'm torn between creating a sample synthesizer, or going with additive synthesis, or physical modeling. Since this is for a practice instrument, not a performance instrument, my goal is a sound that is pleasant, flexible, responsive, and natural, but not necessarily a spot-on concertina impression. The sample-based synth would be the easiest to implement, but the least responsive. Physical modeling would be the most responsive, but so far the physically modeled reed instruments I've heard have sounded a bit sterile. Additive techniques and their offshoots (additive with noise-modulated phase, e.g.) are easy to implement, but painful to parameterize -- requiring extensive analysis of existing instruments. I'm fully prepared to spend hundreds of hours on this, but I don't want to _waste_ hundreds of hours going down the wrong path.

 

So my question is: what synthetic reed instruments have your heard that you liked?

 

Link to comment
Share on other sites

Jim

 

My thinking is to punt that problem down the road and just output class compliant midi over USB. I plan to use the HIDuino library on an Arduino to accomplish this.

 

The concertina sound font that I posted a few days ago is not too bad on a desktop PC running FluidSynth, but my main hope lies with Michael Eskin's samples running on an iPhone, an iPad or an iPod Touch. I understand that they can be accessed through the Thumbjam app.

 

Maybe I am missing something here but an onboard midi synth seems like a lot of work.

 

Don.

Link to comment
Share on other sites

Jim, my 40 key ES midi is up running - so far by normal midi cable or midi to usb.

 

I am planning to use low latency low power bluetooth ( V4.0 ) to send the midi information to an IOS device ( IPad ) or a laptop, which can do all the nasty sound-work! If you have spare hours, you can help me with the programming;-)

Link to comment
Share on other sites

For what it's worth, I've been running a QuNexus (small MIDI keyboard) through my iPhone 6 using the "camera connector kit", a little widget that adapts the USB cable to go into the iPhone's Lightning port. I've been playing the keyboard through ThumbJam, and overall pleased with the program. Some of the sounds, including concertina, are decent, while some are uncannily realistic like sackpipa (Swedish bagpipe).

 

I've tried amplifying my phone through an external Bluetooth spearer, but there's too much lag to play accurately. Though I'm sure a smart person could puzzle the issue out.

 

As more and more folks are owning phones and tablets that can serve as a MIDI device, me personally I don't see motive for the hassle of trying to build the processor into the instrument, but would rather have just the mechanical bits and switches in the instrument, and run a cord out to my phone where all the magic happens.

Link to comment
Share on other sites

Thanks, all. That was very useful. I agree that using a smartphone/iPhone would be a huge time saver, and if I were the only client for this box, that would be a good route. But not all of my dozen children/nieces/nephews have the necessary hardware, and my vision for this is a simple grab-and-go instrument, no more complicated to pick up and play than an acoustic concertina. So my plan is first to make sure the MIDI implementation is solid (I didn't implement MIDI over USB, but I can do it in a future rev). Next, I'll code a simple subtractive synth (a few hours of work), which I'll use to shake out the controller hardware. Lastly, I'll try my hand at a SF2/SFZ sample player, which my hardware may or may not be able to do. If it works, it will give the kids access to a huge base of non-concertina soundfonts to play with. I'll start with Don's soundfont (thanks, Don), and who knows, by the time I get around to it, maybe Michael Eskin will have ported his samples to a format that I can use. I'll triage out additive and physical modeling techinques -- too much of the wrong kind of work for me. And I don't know how much of my work will be useful to you, but you are welcome to all of it. Thanks again.

Link to comment
Share on other sites

I know right now everyone seeing this thread is seeing the Udar thread, but for the sake of folks reading this thread in future months, it's worth checking out the thread on the "Udar" a small vaguely concertina-esque MIDI instrument by a Japanese inventor: http://www.concertina.net/forums/index.php?showtopic=17477

 

Something *very* close to this, just slightly more squeezy and with a Hayden layout, is basically what I was envisioning when I started this thread.

 

5d1lkm.jpg

Link to comment
Share on other sites

I've tried amplifying my phone through an external Bluetooth speaker, but there's too much lag to play accurately. Though I'm sure a smart person could puzzle the issue out.

This is almost certainly the latency lag on the Bluetooth protocols up to V3, which as I understand it had a delay of nearly half a second. I think this was to allow packets of data to be lost in transit and then resent and hence arrive in the wrong order, and be resorted before being played. See the "Complete Assembly" section of the page on my Midi tina.

 

I am planning to use low latency low power bluetooth ( V4.0 ) to send the midi information to an IOS device (IPad) or a laptop, which can do all the nasty sound-work! If you have spare hours, you can help me with the programming;-)

Spare hours are in short supply at present (I'm not yet retired!), but I'd be very interested in your practical findings as to whether Bluetooth V4 devices do avoid the latency referred to above, and also which particular devices and speakers do talk Bluetooth V4.

 

As more and more folks are owning phones and tablets that can serve as a MIDI device, me personally I don't see motive for the hassle of trying to build the processor into the instrument, but would rather have just the mechanical bits and switches in the instrument, and run a cord out to my phone where all the magic happens.

Hear, hear. Separation of action from voice via a (preferably wireless) digital connection is the way to go.

 

Regards,

Link to comment
Share on other sites

  • 2 weeks later...

My first baby is ready! It plays like a dream. I can use it both with an android Nexus 7 and an Ipad, where the operating-voltage is supplied by the tablets! I can play for a couple of hours. And there is no noticeable latency!

 

post-7162-0-34672300-1425850717_thumb.jpg

 

Currently I am, building an anglo. The Hayden will have to wait!

 

I hope to do some recording soon...

Link to comment
Share on other sites

And from the black device Don mentions, the last little white bit leading to your iPad, is that the Apple "camera connector kit" to go from USB to Lightning?

 

 

 

That looks awesome, by the way! Any chance of a YouTube clip showing you setting up and playing your instrument?

Edited by MatthewVanitas
Link to comment
Share on other sites

Currently the concertina outputs normal MIDI. So I converted a MIDI to USB adapter to allow the connection to the pads. Inside the adapter I added the 5V supply to the MIDI cable, which is not standard.

 

On the android pad I need a USB A to micro USB adapter, on the Ipad the "camera-connector" ( it has to be Apple, two cheap versions didn't work ).

Link to comment
Share on other sites

While not an ideal or even very good approximation of what you're going for, I do have both an iPad Hayden-style MIDI controller app available that can be used with a variety of MIDI interfaces to play external synths, or to directly play CoreMIDI compliant synths running on the iPad itself.

 

https://itunes.apple.com/us/app/ijammer-jammer-style-coremidi/id375667582?ls=1&mt=8

 

It's something I did a long time ago in association with Jim Plamondon.

 

 

 

 

 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...