Jump to content

MIOS controlled 16x16 audio matrix


Rowan
 Share

Recommended Posts

Hi

I'm going to build a 16x16 swithing matrix (digitaly controlled pacthbay) which will be controlled by MIOS for a project at university this semester. Origanlly I planned to use relays, but I quickly realised this would cost alot of money (256 relays!!). I've been looking into solid state solutions to this problem and I found a IC which should do the job.

16 input

16 output

256 crosspoint (Any input can be routed to any or all outputs)

Chip can be cascaded to create a Matrix of an size!

The IC is:      AD75019JP

the data sheet is avalible at:

http://www.analog.com/UploadedFiles/Data_Sheets/267321864AD75019_c.pdf

Will I be able to interface this chip with MIOS?

Anyone have any better ideas to achive the same result?

Anyone interested in developing this with me so we can release this to the MIDIbox community?

This IS going to happen, so help on ANY level would be useful.

I am currently developing the spec. so input at this stage will be very helpful.

Feedback please

Rowan

Link to comment
Share on other sites

DUDE!!  ;D ;D ;D  NICE!!

How did you find that chip!?! I had this same idea a few months back and gave up because of the whole relay thing... Oh and BTW... Balanced audio is 3 core cable so it's not 256 relays its 768..eep... (I dunno maybe they could share an earth and it's 512...whatever, it's a lot of relays!)

Man, great idea, great work finding the chip... I don't know much about electronics as such, but programming I'm cool with... So any help I can give, is all yours.

I look forward to seeing this thing finished :) I might even do a larger version (say 32x32 or 48x48) and never touch an audio jack again  ;D

Cheers!

Link to comment
Share on other sites

Hi

Here is my initial idea for the spec:

Any input can be routed to any output.

inputs and outputs can be ganged.

Display (LED,LCD etc) to show which input is routed to which output .

Ability to store set-up's

Mains powered (No wall wart!!).

Clip LED's on the input's.

Balanced I/O.

+4dBu Op level.

NO fan.

Removable front panel for remote editing.

Anymore ideas that are realisitic?

Rowan

Link to comment
Share on other sites

great idea!

I think it can work with mios with some program mod.

this is maybe the solution for the mixer I plan to build...

I'm not sure, but the only problem is if for example you plug input 1 to output 1 and 2, and input 2 to output 2 and 3, this will route input to output 3 and input 2 to output 1...

Link to comment
Share on other sites

Hi

Does anyone know what rate the 18F transmits data to the shift registers in the DOUT modual?

The IC I'm looking at using reqiures data to be clocked between 20kHz -> 5MHz. The IC have 2 data input lines, 1 for serial data transfer from the Micro to the first set of internal shift registers, and a second parellel input to move the 256 bits of data recived to the next block of memory.  As the data is transfered in serial to the register I assume that the parallel input should be running at 1/256th the speed of the serial input.

Any ideas?

Rowan

Link to comment
Share on other sites

Hi

Thanks Ian, I might have to take you up on the offer. I'm getting nowhere fast trying to figuer out how to get MIOS to talk to the chip. Could you give me the part number for the relays so I can check out the specs

Thanks

Rowan

Link to comment
Share on other sites

Don't give up Rowan, I'm sure that this can be done with that IC... Flownez and I discussed it at length this weekend, in fact he pointed out that if you didn't nheed MIDI control you wouldn't even need MIOS, you could just run the thing through a serial port...

Anyway I want MIDI control and I'm keen to do this, the cheap way with the IC...

I guess the real problem is that I don't have heaps of time on my hands, I work pretty long hours... I know this is a university program for you so it's not something that you can just do "whenever"... However I won't be doing much more work on the sequencer until TK releases the 18F port of the MB64Seq, so maybe between the three of us...?

What's your timeframe look like?

Anyway as for clocking... now before I open my big mouth I'd like to reinstate my previous comments about how I'm sh*t with electronics ;)

According to the datasheet:

The serial shift register is dynamic, so there is a minimum clock rate of 20 kHz. The maximum clock

rate of 5 MHz allows loading times as short as 52 ms.

 

From that statement and the other nerdy stuff on that doc, I've gathered that basically you push data to control the switches into the serial in, and push it a clock signal after whenever the serial data for each switch is loaded. This shifts the data to the next register... Once all the required registers are full (you've finished sending it data about how to configure the switch), you send it a PCLK pulse and that moves all the data about how to set the 256 switches, into the chip and moves the switches accordingly...

Provided of course that the data you push into it doesn't clock below 20khz which is the minimum clock speed...

I should have read this datasheet closer before :)

So the MIOS side of it will need three digital outs... One to push data to the chip which represents the state of the switches. One to push a clock signal to lock those bits in place. And one to send the parallel clock signal to tell the chip 'OK go load this stuff into the switches and patch this sucker already'...

Food for thought? Geez I hope I got this techy crud right or I'll look like a right dork!

Anyway here's the juicey bit from the datasheet:

APPLICATIONS INFORMATION

Loading Data

Data to control the switches is clocked serially into a 256-bit shift register and then transferred in parallel to 256 bits of mem-ory. The rising edge of SCLK, the serial clock input, loads data into the shift register. The first bit loaded via SIN, the serial data input, controls the switch at the intersection of row Y15 and column X15. The next bits control the remaining columns (down to X0) of row Y15, and are followed by the bits for row Y14, and so on down to the data for the switch at the intersec-tion of row Y0 and column X0. The shift register is dynamic, so there is a minimum clock rate, specified as 20 kHz. After the shift register is filled with the new 256 bits of control data, PCLK is activated (pulsed low) to transfer the data to the parallel latches. Since the shift register is dynamic, there is a maximum time delay specified before the data is lost: PCLK must be activated and brought back high within 5 ms after fill-ing

the shift register. The switch control latches are static and will hold their data as long as power is applied.

To extend the number of switches in the array, you may cascade multiple AD75019s. The SOUT output is the end of the shift register, and may be directly connected to the SIN input of the next AD75019.

Link to comment
Share on other sites

Hi

Ian: Thanks, I'll give you a call when I've got a chance to arrange things.

stryd_one: Cool, I'm pleased you are still interested. I'm planning on persuing both options at this stage (IC and Realys). I've got to keep my options open to the possible solutions to this problem,  My understanding of how the chip will work was pretty similer to yours.

I'm going to order the IC this week to breadboard the circuit. I can take care of all the electronic prototyping, PCB design etc. What I really need is someone to write the code for MISO.

Anyway, I've got a little over three months to complet this design.

I'm thinking of using two cores for this project, 1 for the control for the switching matrix and a second core for the remote. The reason i've chosen to use a core in each unit is so the remote does not have to be connected for the unit to function. This useful for set and forget situations where more than one unit is in use. The idea I have for the remote is simple and funtional:

40x2 LDC with 8 pushbutton encoders below

4 Menu navigation buttons

bank sav/select buttons

MIDI out

I'm going to redesing the PCB's so that all compnents fit on one board to minimise size and wiring. It is possible I may use surface mount components.

Enough talk, I've got to get my hands on on of the IC's and relays to get to work

later

Rowan                

Link to comment
Share on other sites

  • 2 weeks later...

Hi

Thank for pointing out the IC illiminator. I've looked at the data sheet and the chip looks more suitable than the  AD75019JP. The data sheet provides good application notes, which are always helpful.

How did you go about ordering samples from Analog Devices. I thought that samples where not avalible for these IC's.

The Chip is quite expensive though:

$44.54 from Newark inone. part number is 33C3515

Thanks

Rowan

Link to comment
Share on other sites

Easy.

I asked customer support via email about samples, stating that there was no way to add the device to the sample cart. Right, they said, currently there is no automated way of getting samples, but: I should contact the local AD representative, they could arrange for samples. That's what I did, about two weeks later I got both samples.

Regards, ilmenator

Link to comment
Share on other sites

Hi

I've just been in contact with Analog Devices in the US of A. I'm glad to say that they were more than happy to send samples of the AD8113JST, even after I told them it was for a non-comercial product. :)

Good on you Analog Devices!!!! :) :)

Just about time to turn all this talk into action.

Rowan

Link to comment
Share on other sites

Hi

I've just found another product that can be used for matrix type switching and Alot more (Almost all dynamic processer) This product is designed specifically for  pro audio apps, which is a big bonus.

When used in switching apps it will be able to interface directly with the DOUT modual. As far as MIOS is concerend very little code should  have to be changed in order to get it to work.

Two devices per "switch point"will be needed to achive good attenuation of signals when "off", so this device would not be pratical for large scale switching. But would be fine of small projects.

The device can be found at:

http://www.silonex.com/audiohm/index.html

I haven't found any prices yet, though the manufacture claims they cost less than VCA's

I don't know if anybody has worked out quite where I'de like all this dicussion to lead to.

What I'm working towards (apart from a project for uni) is an analog mointor section for a MB which is under the control of MIOS (poss. more analog features which are of use to studio based MB users).

The Mackie HUI has such a monitor section. If you would like to know more about whan I'm talking about.

Rowan

Link to comment
Share on other sites

Hi

Once again I've got another possible solution to the switching problem. Once again Analog Devices have come up with the goods.

The IC is: SSM2412

It's designed for Audio (Cool), has excellent THD and channel isolation figures and can handle a massive +30dBu. And direct interfacing with the DOUT.

This IC is look very good for small scale switching.

I've almost finished the inital report on the project (approx 2500 words at this stage) which I'll try to post here if it is not to big (what is the limit on the size of a  post???).

Till next time

Rowan

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...