Track display to rotating leds

[ris8_allo_zen0]

Hello,

since a few days I was hit by an idea: build a rotating LED bar for a persistence-of-vision display like SpokePOV, and use it as a pattern display for a SEQ.

Each orbit represents a track and its steps are equally distributed into its length, according to the track length.

All this can help to have a global view of the tracks, but also to get a nice light effect for the public during live playing! :D

SpokePOV, among other similar projects (see also "propeller clock"), is made of a LED bar driven by some shift registers which are in turn driven by a microcontroller that has the image bitmap and knowledge about the rotational speed and phase. The bitmap is usually static or dynamically generated (e.g. by current date&time), the rotational data are read by a sensor that triggers when a revolution is made.

All of this stuff must reside into the rotor / wheel / fan / anything able to spin. The difficult part is how to interface this circuitry with the rest of the world, since physical connection is not so easy and IMHO a wireless solution is more efficient and elegant.

I thought about an IR interface between the SEQ core and the rotating microcontroller. The former sends the updates of the tracks (step activation, track length, cursor position) by some ad-hoc protocol, and the latter reads these data and paints the air accordingly :)

An IR interface also allows me to use the POV display for other purposes by simply using an IR remote control.

According to this page, realizing a simple IR transmitter only requires two pins of a PIC, whose one is a PWM output calibrated to ca. 38kHz.

What I'm asking you, is what do you think about this idea and if it's possible to implement such a IR transmitter into the SEQ hardware and software.

Many thanks!!

[stryd_one]

I think it's a great idea. I had this idea too when thinking about how to represent tracks of different lengths that loop in the same time... IE, 7 buttons on a CS that loop in the same time as 4.... But I gave up when trying to find a way to send live data to it. Good luck with the IR transciever! I suspect you'd have to write a whole new CS, but I'm sure it's doable.

[Sasha]

I love POV and propeller clocks. Make sure you watch POV rotating globe at youtube. Absolutely stunning.

I think POV display would be cool addition to SEQ, but same time I think it would not be very comfortable to use... buttons in line and their representation in circles. :-\  Anyway, I`d like to build it.  ;)

[stryd_one]

buttons in line and their representation in circles. :-\

Well I would have the tracks run around the circumference of the circle, with step 1 in the vertical position, so a track will really "loop" :)

[Sasha]

Well I would have the tracks run around the circumference of the circle, with step 1 in the vertical position, so a track will really "loop" :)

I was thinking of that too inspired by Revolution. And even it is true loop representation in-line can be much more compact. I doubt there are already some circular SEQs around, and common in line wouldn`t match the POV display well.

[ris8_allo_zen0]

My aim was not to make a CS out of a POV display, but simply to design an output-only interface used in addition to the standard CS. It will provide a global view of the patterns, or simply something nice to see while playing or dancing.

But for you that think about an input too, I have a veeery conceptual idea: a "ray bar" in front of the POV wheel, attached to the center by a rotary encoder and having a row of 16 buttons and leds that overlap their POV counterparts. The bar can be freely moved around the circumference to select steps, and with buttons you change the steps in each orbit/track.

Another idea: a fixed ray bar and a knob in the center. The knob rotates the POV disc: spin it until you see your steps under the bar (and re-represented by the Leds in the bar), then edit accordingly.

[stryd_one]

My aim was not to make a CS out of a POV display, but simply to design an output-only interface used in addition to the standard CS.

Heh, until you posted those last two ideas, I wouldn't have said a CS based on the POV display would be practical ;)

[cimo]

..why the loop/seq representation has to be circular? can t it be linear as in a normal sequencers ?

and, lately i ve seen quite a lot of spare touchscreens for hand held devices sold on ebay for cheap... some of them has a dual analog output based on resistance ie: the position can be detected by 2 resistance values on the 2 axis.That could be easily interfaced to  a couple of AINs.

just a thought..

simone

EDIT: about the wireless transmission of data i can t understand how you think you could use IR.Isn t a radio transmission required in this case?

[ris8_allo_zen0]

..why the loop/seq representation has to be circular? can t it be linear as in a normal sequencers ?

Simply because I imagine a loop as something circular, and maybe you do too :)

and, lately i ve seen quite a lot of spare touchscreens for hand held devices sold on ebay for cheap... some of them has a dual analog output based on resistance ie: the position can be detected by 2 resistance values on the 2 axis.That could be easily interfaced to  a couple of AINs.

Good to know! But I think this POV display should be quite bigger than a handheld display. Think about a room fan, or the device Sasha reported here.

Another idea (please forgive me!....): instead of each button, a pair of IR led and receiver that can sense the approach of a finger near them. They could be put together to the leds in the POV bar. That makes a touchscreen POV display!

EDIT: about the wireless transmission of data i can t understand how you think you could use IR.Isn t a radio transmission required in this case?

Why should a radio means required for this application? Wireless means "without wires", not strictly radio communication :)

[stryd_one]

Simply because I imagine a loop as something circular, and maybe you do too :)

Actually I like to think of it as a spiral, but I like to change my music a lot as it plays ;)

a pair of IR led and receiver that can sense the approach of a finger near them. They could be put together to the leds in the POV bar. That makes a touchscreen POV display!

OK, stop it. I have enough projects already  ;D

[cimo]

Quote from: cimo on Today at 11:03

..why the loop/seq representation has to be circular? can t it be linear as in a normal sequencers ?

Simply because I imagine a loop as something circular, and maybe you do too Smiley

ok so you do want it circular! up to taste!

Quote

EDIT: about the wireless transmission of data i can t understand how you think you could use IR.Isn t a radio transmission required in this case?

Why should a radio means required for this application? Wireless means "without wires", not strictly radio communication Smiley

cause an IR communication needs the receiver and the transmitter to be in sight of each other.

An alternative would be having a crown of receiving leds spinning around while a fixed transmitter sends an interrupted data stream, all synched to the whole system..

[ris8_allo_zen0]

cause an IR communication needs the receiver and the transmitter to be in sight of each other.

That's not a big problem for me. What I want is to make this thing compatible with good old IR remote controls, so I can use it for other purposes while not using the SEQ.

Choosing a radio interface instead of an IR one is not a major issue. It can be an alternative, though.

An alternative would be having a crown of receiving leds spinning around while a fixed transmitter sends an interrupted data stream, all synched to the whole system..

Do you mean "interleaved" data stream? That would need N infrared receivers, each with its own circuit to drive a Led. Since there's already a microcontroller, I think it's better to attach just one infrared receiver to it, then write some protocol in software.

Please be aware that these are just fuzzy ideas, not final decisions. I write them here hoping they'll get useful to somebody (edit: forgot to complete the sentence :D)

[TK.]

What I'm asking you, is what do you think about this idea and if it's possible to implement such a IR transmitter into the SEQ hardware and software.

See MBSID firmware, USER_Init hook how to enable PWM (in this firmware it's used to provide the 1 MHz SID clock). For 38 KHz clock, you want to use the same configuration as shown in the Basic Code example.

Transmitting informations at 1kbaud should be easy, because you could control the NAND from USER_SR_Service_Prepare or USER_SR_Service_Finish. These hooks are called each mS

Using different baudrates is impossible, because all other timer sources are already allocated by the firmware.

Best Regards, Thorsten.

[cimo]

Quote from: cimo on Yesterday at 13:41

cause an IR communication needs the receiver and the transmitter to be in sight of each other.

That's not a big problem for me.

maybe i was not clear enough, my point is that if the whole thing spins around, how do you think you will send any infrared data?

aside from the "leds crown" idea, it could be achieved with a central photoled/led thingy

[ris8_allo_zen0]

Transmitting informations at 1kbaud should be easy, because you could control the NAND from USER_SR_Service_Prepare or USER_SR_Service_Finish.

That's fine! It's a good point to start.

maybe i was not clear enough, my point is that if the whole thing spins around, how do you think you will send any infrared data?

aside from the "leds crown" idea, it could be achieved with a central photoled/led thingy

Sorry for not immediately understanding that (I also misread crown as crowd)! The IR receiver should be placed as near to the hub as possible, along with other components so that the momentum doesn't rise too much. Anyway an IR receiver has usually a very high viewing angle (110° or so), so as long as the transmitter is positioned in front of the wheel, there shouldn't be any problems.

[cimo]

Anyway an IR receiver has usually a very high viewing angle (110° or so), so as long as the transmitter is positioned in front of the wheel, there shouldn't be any problems.

what about your body/hands? i think this part is more delicate at least if we are talking about a UI and not just a "show" thingy for the public.

Best thing would probably be using either a belt drive engine or a small electric motor and the receiver photo led in the back side

Good to know! But I think this POV display should be quite bigger than a handheld display. Think about a room fan, or the device Sasha reported here.

yes i agree on that, unfortunately bigger touchscreens are capacitance based but probably something can be found.

simone

[DrBunsen]

Best thing would probably be using either a belt drive engine or a small electric motor

What about a discarded record player?

[cimo]

Quote from: cimo on 2008-03-04 at 11:12

Best thing would probably be using either a belt drive engine or a small electric motor

What about a discarded record player?

too slow, you would need to replace the engine, but it s a good idea for the mechanical construction