Pandora as infrared universal remote control?


HelenF

Very Active Member
Joined
Jun 22, 2013
Messages
615
Location
UK
I'm trying to think of a way to drive a bright IR LED with raw remote control signals, while keeping the whole thing as a practical handheld device.

I currently have something working with an Arduino on the OTG port (with a big capacitor etc for the LED), so the easiest thing might be to mount the Pandora and the Arduino together somehow. I'd rather not make any permanent changes to the Pandora to achieve this.

The neatest way would be a tiny device that fitted directly into one of the three ports on the back - might there be some existing product for the purpose? With USB, I don't think a home-made product could be small enough. EXT might be possible.

Has anyone done something like this before? Any suggestions?
 
Yes, a solution involving a circuit board on the other end of a USB cable is fine for a receiver.

Questions on the EXT connector:

http://pandorawiki.org/File:Cable_connector_pinout.png

This diagram says Power Out is "1.5V-3.0V @ 200mA" - why the variation in voltage? Also, what voltage and current do the PWM pins provide?
 
Hmm, that's quite a range. Seems to have been pulled directly from the hardware hacker's guide, which AFAIK hasn't been updated in a while. It could conceivably be a scaled battery voltage, but I don't know why you'd have that available on an external interface like that - I'd expect either raw battery voltage or something regulated.


If you wire up something to the ext connector, it'd be nice to know what voltage you do get here, and through the PWM pins. I never got round to soldering up the ext connector I've got.
 
IIRC the voltage on the ext connector is regulated, and the voltage it's regulated to can be set in software.

I also recall that it resets to the default voltage on reboot though, so using a different voltage level to the default that was used for the original dev breakout boards could be problematic if the connected device is still connected when the Pandora reboots.

- Neelix
 
UART3 can be configured in IR mode. Just connect an IR LED and resistor to it and bang your bits out on /dev/ttyO3. Use ctl commands to change the bitrate and all that familiar jazz. Put an IR receiver on the RX line and get a universal remote reader as well.
 
According to this page, the default power supply is 2.8V.

http://pandorawiki.org/Configuring_ext_signals

The "Pandora Hacker Guide" PDF says, "All digital pins on this connector are 1.8V logic and must be level shifted if they need to be interfaced to a higher/lower voltage system."

For receiving, a typical IR demodulator won't work with a power supply as low as 1.8V, and it would output at its power supply voltage, so additional circuitry would be required to connect one of those.

So far, I've not been able to find anything about the maximum current output from the digital pins. (But I'd be surprised if it was more than 20mA.)

Someone is saying that OMAP3 can generate Consumer IR signals, but they go out on the uart3_cts_rctx pin, while IrDA goes out on uart3_tx_irtx. Is this something different from what you're talking about WizardStan?

http://lists.goldelico.com/pipermail/gta04-owner/2013-January/003772.html
 
Oh, hmmm... I'm probably mistaken and have been giving bad advice.

edit: wait, no, I was right!

GPIO_165 and _166, which are the pins exposed on the EXT port, are the uart3_rx_irrx and _irtx pins.

Yes, I think that is exactly what I was talking about, using gpio_166 to drive the IR LED directly.
 
Last edited by a moderator:
According to the email I linked, _irtx is for IrDA, which is no use for remote control; while the remote control pin is _rctx (and it doesn't seem to be on the Pandora EXT connector).

Other possibilities mentioned in that email thread were using GPIO bitbang; PWM; or a UART set to a 38kHz bit rate and sending bytes related to the IR pulses rather than the real data bits.

And I was saying I doubt the IO pin can drive the IR LED directly because of current requirements. But 200mA from the power pin is fine, so it would be simple enough to switch a FET connected to that.
 
Last edited by a moderator:
Oh I see. Dang, I misunderstood what irda was then. Sorry. :(

I just thought it was UART serial interface, didn't occur to me that there'd be layers on top of that.
 
Strangely, on the EXT connector page in the hackers manual PDF is written "UART3 can also be configured for IrDA (Infrared Data Association) or CIR (Consumer Infrared Modes)". Technically, this doesn't contradict the above email, but the implication does ;)
 
The EXT connector is smaller than I remember. How do you solder these things?

It seems there were a limited number of "TinyBoB" made by a community member, but nobody has any extra ones.
http://boards.openpandora.org/topic/13461-spare-tinybob//URL]

I'm surprised there was no official product. It would be a useful thing for ED to stock. EXT connector + TinyBoB already soldered would be even better for those of us without much of a steady hand :)
 
Last edited by a moderator:
The EXT connector is smaller than I remember. How do you solder these things?

It seems there were a limited number of "TinyBoB" made by a community member, but nobody has any extra ones.

http://boards.openpandora.org/topic/13461-spare-tinybob//URL]

I'm surprised there was no official product. It would be a useful thing for ED to stock. EXT connector + TinyBoB already soldered would be even better for those of us without much of a steady hand :)
Yeah, the EXT connector are a bitch to solder.. The TinyBoB was made by a community member peca who hasn't been around in ages. The Tinybobs were great, but need something to keep them from flexing. On two occasions just general use caused them to rip out the pins on the plug side. Too bad I used the four of TinyBoBs I had. About the pins being pulled out issue, I did see this on thingiverse for the EXT connector for those with a 3D printer, but not sure if it's tinyBoB friendly or not. : http://www.thingiverse.com/thing:34176
 
Last edited by a moderator:
Once it's soldered, would it work to just cover the joint with glue?

I wonder how hard it would be to solder a different USB connector onto one of these (smallest FTDI board). I think the spacing is smaller than the EXT connector, but there's only 5 lines and they're on a PCB already. And these things are a bit cheaper than EXT connectors if I wreck a few.
http://www.ebay.co.uk/itm/251353615134

Could maybe use a connector with wires, and glue it sideways to the bottom of the board.
 
Last edited by a moderator:
Once it's soldered, would it work to just cover the joint with glue?
Well I did use hot glue, but small flexing of the board still caused the pins to break after a while. It may not be as much of an issue if you just solder directly to the pins and find a way to keep the wires from moving. It's just hard to solder directly to those pins.
 
That would work, but it's a technically unsatisfying solution for a computer engineer as it ties up both the headphone port and the volume wheel, and goes analogue prematurely. The software should be easier for a digital solution too IMO.


Regarding the EXT connector, I've measured mine to have a vertical pitch of pretty much damn on 1.5mm, meaning a common 1.5mm PCB should be able to slot in between the pins (assuming the traces aren't too thick, I guess). Furthermore it's 8.1mm between the two stress relief tabs, so you need an edge on your 1.5mm thick PCB 8.1mm wide and at least 1.2mm proud of the main edge and it should fill the gap, and if you solder the strain relief tabs and hot glue to whole thing, I'd hope you'd end up with something that is solid enough, especially when going to something lightweight like an LED.


I make the leads themselves to be .3mm wide (and around .2mm tall - they're rectangular by eye) and are spaced at 1.0mm intervals, including at the end (where the strain relief tabs are). There's 7 pins, so 8 intervals. Technically that adds up to 9.8mm so it looks like I was bending them slightly when measuring. I've just taken an external pitch measurement (2x pin width plus one interval) and get 1.2mm. I suspect the natural pitch is close to 1mm, with the pins closer to .2mm wide (and the gap being .8mm).
 
I did something once with digital signals on headphone output (not for IR). It wasn't very reliable, and handling the wave data on the software side is not nice. I'd also be concerned about current requirements again; is it safe to connect the LED directly like that? and there was someone in the comments saying the LED wasn't bright enough.

Trial-and-error suggests my LED needs about 150mA for good performance as a "remote" control.

I wasn't planning to get into the business of making PCBs. But it might come to that at some point, if not for this project then for a different one ;)
 
I did something once with digital signals on headphone output (not for IR). It wasn't very reliable, and handling the wave data on the software side is not nice. I'd also be concerned about current requirements again; is it safe to connect the LED directly like that? and there was someone in the comments saying the LED wasn't bright enough.


Trial-and-error suggests my LED needs about 150mA for good performance as a "remote" control.


I wasn't planning to get into the business of making PCBs. But it might come to that at some point, if not for this project then for a different one ;)
I did a crude fiber optics project for a science fair when I was in high school.. I needed to drive the IR LED with a small transistor and an external power source to properly drive it.. I had another crude amplifier on the receiving end with a speaker.. It sounded not too bad, ended up going to the state science fair, for a project that took me all of an hour to put together. I guess fiber optics was new and amazing back in 1992
 
Last edited by a moderator:
Back
Top