Universal IR remote controller

[Thomas W]

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on the
Web but didn't find all I need. I wonder especially how the device will be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the project
for Monday.

Thanks in advance,
Thomas W.

 

[martin griffith]

On Fri, 27 May 2005 14:33:10 +0200, in sci.electronics.design "Thomas
W" <thomas_w@op.pl> wrote:

Hi All,
I need to design an universal IR remote controller for my university
project.

I'm pretty desperate since I have to finish the project
for Monday.

Thanks in advance,
Thomas W.

I think you have left it a bit late so

Rip the chip out of a 1 for all remote, put it in a die cast box.

http://www.phdcomics.com/comics/archive.php?comicid=221



martin

"Wales is a big welsh-shaped rain collection device"

 

[Roger Hamlett]

"Thomas W" <thomas_w@op.pl> wrote in message
news:d77488$1uq$1@news.onet.pl...

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on
the
Web but didn't find all I need. I wonder especially how the device will
be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the
project
for Monday.

Thanks in advance,
Thomas W.
You don't.

Build an IR detector, and sampling system, that can read the IR data, at a
significant multiple of the highest expected signal rate. Have a button to
trigger 'learn', and have the system start recording when it sees IR
activity. Store the sequence for a period of time. Then to simulate that
command just replay the data pattern to an IR driver. So long as the
period stored is longer than the expected maximum signal length, and the
sample rate is high enough, you can simulate any remote.
You don't store the 'signal codes', but record the IR pattern being
produced.

Best Wishes

 

[Mebart]

In these days of inexpensive PIC's and small non-volitile ram
availability, you are lucky. The parts and the expertise to do the job
is easily and cheaply attainable.

You need a PIC, a small non volitile ram chip and an inexpensive
(short range) IR transceiver. With proper programming, this is a
somewhat basic project. You might find that someone has done it
already and/or has written the code that you can tweak to make it
happen in your own case.

Check out the piclist mailing list at mit.edu. The mailing list is
very active and the group has a large searchable archive of messages
and many mb of searchable projects on the server.

Regards,

Mebart

On Fri, 27 May 2005 14:33:10 +0200, "Thomas W" <thomas_w@op.pl> wrote:

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on the
Web but didn't find all I need. I wonder especially how the device will be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the project
for Monday.

Thanks in advance,
Thomas W.

 

[Thomas W]

Użytkownik "Roger Hamlett" <rogerspamignored@ttelmah.demon.co.uk> napisał w
wiadomości news:LwFle.314$ci4.20@newsfe6-win.ntli.net...

You don't.
Build an IR detector, and sampling system, that can read the IR data, at a
significant multiple of the highest expected signal rate. Have a button to
trigger 'learn', and have the system start recording when it sees IR
activity. Store the sequence for a period of time. Then to simulate that
command just replay the data pattern to an IR driver. So long as the
period stored is longer than the expected maximum signal length, and the
sample rate is high enough, you can simulate any remote.
You don't store the 'signal codes', but record the IR pattern being
produced.

Best Wishes

Thanks, that seems to be a very good idea. However I'm still wondering how
do you imagine recording the pattern...
I imagine it that way: The IR signal for remotes is usually sent with freq.
38kHz, and different codes have different periods of the low/high levels,
thus I think after detecting a signal I should check it every 26us and
increment a counter each time. When the signal inverts, I store the value of
counter, null the counter and repeat the procedure until reaching the longer
silence (like 20 ms). I calculated that to store all the possible codes
(that I know) it will be necessary to have 512 bits of memory per each
code/button. (Maximum of 32 signal inverts times 16 bits for the counter
value)
Is the above algorithm reasonable or can you think of any more effective
ones?

Thanks a lot

 

[Luhan Monat]

Thomas W wrote:

Użytkownik "Roger Hamlett" <rogerspamignored@ttelmah.demon.co.uk> napisał w
wiadomości news:LwFle.314$ci4.20@newsfe6-win.ntli.net...


You don't.
Build an IR detector, and sampling system, that can read the IR data, at a
significant multiple of the highest expected signal rate. Have a button to
trigger 'learn', and have the system start recording when it sees IR
activity. Store the sequence for a period of time. Then to simulate that
command just replay the data pattern to an IR driver. So long as the
period stored is longer than the expected maximum signal length, and the
sample rate is high enough, you can simulate any remote.
You don't store the 'signal codes', but record the IR pattern being
produced.

Best Wishes




Thanks, that seems to be a very good idea. However I'm still wondering how
do you imagine recording the pattern...
I imagine it that way: The IR signal for remotes is usually sent with freq.
38kHz, and different codes have different periods of the low/high levels,
thus I think after detecting a signal I should check it every 26us and
increment a counter each time. When the signal inverts, I store the value of
counter, null the counter and repeat the procedure until reaching the longer
silence (like 20 ms). I calculated that to store all the possible codes
(that I know) it will be necessary to have 512 bits of memory per each
code/button. (Maximum of 32 signal inverts times 16 bits for the counter
value)
Is the above algorithm reasonable or can you think of any more effective
ones?

Thanks a lot

The first thing I would do is get a commercial universal remote, bench
wire an IR receiver module to a digital scope, and look at a lot of the
coding patterns. Then imagine a good way to store all of those. This
is exactly what the engineers that produced the commercial unit had to
do to make it in the first place -- and their design was very cost
sensitive.

--
Luhan Monat: luhanis(at)yahoo(dot)com
http://members.cox.net/berniekm
"Any sufficiently advanced magick is
indistinguishable from technology."

 

[Roger Hamlett]

"Thomas W" <thomas_w@op.pl> wrote in message
news:d77cg8$e5q$1@news.onet.pl...

Użytkownik "Roger Hamlett" <rogerspamignored@ttelmah.demon.co.uk
napisał w
wiadomości news:LwFle.314$ci4.20@newsfe6-win.ntli.net...

You don't.
Build an IR detector, and sampling system, that can read the IR data,
at a
significant multiple of the highest expected signal rate. Have a button
to
trigger 'learn', and have the system start recording when it sees IR
activity. Store the sequence for a period of time. Then to simulate
that
command just replay the data pattern to an IR driver. So long as the
period stored is longer than the expected maximum signal length, and
the
sample rate is high enough, you can simulate any remote.
You don't store the 'signal codes', but record the IR pattern being
produced.

Best Wishes



Thanks, that seems to be a very good idea. However I'm still wondering
how
do you imagine recording the pattern...
I imagine it that way: The IR signal for remotes is usually sent with
freq.
38kHz, and different codes have different periods of the low/high
levels,
thus I think after detecting a signal I should check it every 26us and
increment a counter each time. When the signal inverts, I store the
value of
counter, null the counter and repeat the procedure until reaching the
longer
silence (like 20 ms). I calculated that to store all the possible codes
(that I know) it will be necessary to have 512 bits of memory per each
code/button. (Maximum of 32 signal inverts times 16 bits for the counter
value)
Is the above algorithm reasonable or can you think of any more effective
ones?

Thanks a lot
Realistically, a processor will be cheaper and simpler than using

counters. Timing the 'on' pulses using a tone detector, is a pretty
economical way to store the data, but does not cope with dealing with two
or more different clock rates (some rare systems use 44kHz signalling
rather than 38kHz). The commercial units using the sampling approach, just
buffer the output of the photodiode, and use a delta ADC, to just return a
bit stream reflecting whether the signal has increased/decreased from the
last sample. Look at the codes for Bang and Olufsen remotes, if you want
to add some headaches!...

Best Wishes

 

[Rich Grise]

On Fri, 27 May 2005 14:33:10 +0200, Thomas W wrote:

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on the
Web but didn't find all I need. I wonder especially how the device will be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the project
for Monday.

LOL!

Thanks!
Rich

 

[Rich Grise]

On Fri, 27 May 2005 08:06:05 -0700, Luhan Monat wrote:

Thomas W wrote:
Thanks, that seems to be a very good idea. However I'm still wondering how
do you imagine recording the pattern...

The first thing I would do is get a commercial universal remote, bench
wire an IR receiver module to a digital scope, and look at a lot of the
coding patterns. Then imagine a good way to store all of those. This
is exactly what the engineers that produced the commercial unit had to
do to make it in the first place -- and their design was very cost
sensitive.

I worked at a place that made one of the first ones of these things -
"Universal Electronics", I think it was. They had a huge pile of all
sorts of remotes that people had sent them from around the world,
and there was one guy whose job, essentially, was to sit there and
capture codes from the various remotes. Full-time. ;-)

Cheers!
Rich

 

[Mac]

On Fri, 27 May 2005 14:33:10 +0200, Thomas W wrote:

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on the
Web but didn't find all I need. I wonder especially how the device will be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the project
for Monday.

Thanks in advance,
Thomas W.

If you really have to finish by Monday (that is, with a working
remote) you are totally screwed. Take today to contact your
professor and/or file paperwork for an incomplete, if possible. Then go
see Revenge of the Sith, or whatever, and have a good time.

Or go study for your other classes, if needed.

This may sound harsh, but I am honestly trying to help you. ;-)

If all you have to submit is a design sketch or plan, then you might have
a chance, and in that case carry on.

--Mac

 

[Luhan Monat]

Mac wrote:

On Fri, 27 May 2005 14:33:10 +0200, Thomas W wrote:


Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on the
Web but didn't find all I need. I wonder especially how the device will be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the project
for Monday.

Thanks in advance,
Thomas W.


If you really have to finish by Monday (that is, with a working
remote) you are totally screwed. Take today to contact your
professor and/or file paperwork for an incomplete, if possible. Then go
see Revenge of the Sith, or whatever, and have a good time.

Or go study for your other classes, if needed.

This may sound harsh, but I am honestly trying to help you. ;-)

If all you have to submit is a design sketch or plan, then you might have
a chance, and in that case carry on.

--Mac

Show a circuit with an IR receiver module going into a micro controller,
show the micro controller driving an IR LED. Show a matrix of buttons
going to the micro controller.

Tell your professor "Its all done except for the software".

--
Luhan Monat: luhanis(at)yahoo(dot)com
http://members.cox.net/berniekm
"Any sufficiently advanced magick is
indistinguishable from technology."

 

[Mebart]

I confess I started reading this thread because of our own need for a
universal remote control. We have (so far) purchased 3 'all in one'
remotes to replace the aging units that came with the tv and vcr. Each
one works fine...except that some functions are missing and there are
flaws in each design. They all change channels, do basic recording
etc, but try to set up a timed record event and there is no key to get
into the vcr's set up. So, they are almost useless, for lack of 1 or 2
features that seem to be 'left out'.

I had wondered if they still made a remote that would read codes from
the existing stock remote unit. I know they did this in the early
days, but lately there was no need because non-volitile ram and
EEPROM's could hold data for hundreds of different home entertainment
devices.

Lately, there are new models coming out and the manufacturers can't
keep up...........so, there seems to be a resurgence in the
manufacture of 'learning' remotes.

I looked on ebay and there are many models that advertise that they
'learn' the code from the actual remote control to be cloned. A few
sport backlit screens and USB connectors so they can be updated over
the internet! I checked out a few of these in detail, and found that
they all rely on the basic codes stored in eprom and you have to
select what code you need for your particular device. When you run
into a function that you can't use for whatever reason, you have extra
buttons that can be customized (based on the output of the original
remote).

I was not clear whether you could roll your own completely from
scratch, or whether you had to start from the basic codes stored in
the remotes eeprom.

But, clearly....this is the wave of the future. We intend to purchase
one of these units soon.

While this can't help the original ops question regarding how to
design one....perhaps he could purchase one and dissect it to get a
starting place and go forward from there with his own design.

Regards,

M

On Fri, 27 May 2005 14:33:10 +0200, "Thomas W" <thomas_w@op.pl> wrote:

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on the
Web but didn't find all I need. I wonder especially how the device will be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the project
for Monday.

Thanks in advance,
Thomas W.

 

[Don Baker]

I bought an RCA RCU807 unit from overstock.com. This unit has a learn mode
that copies command from an existing remote and cost $25. Hope that helps.
The high end product is the sunwave unit. It has not only learn and macro
(combined commands for multiple devices) but is PC programmable. Check it
out at http://www.remotecentral.com

-=Don=-


<Mebart> wrote in message news:nasg915fr9tfle9epb44a8ivts81qsacbf@4ax.com...

I confess I started reading this thread because of our own need for a
universal remote control. We have (so far) purchased 3 'all in one'
remotes to replace the aging units that came with the tv and vcr. Each
one works fine...except that some functions are missing and there are
flaws in each design. They all change channels, do basic recording
etc, but try to set up a timed record event and there is no key to get
into the vcr's set up. So, they are almost useless, for lack of 1 or 2
features that seem to be 'left out'.

I had wondered if they still made a remote that would read codes from
the existing stock remote unit. I know they did this in the early
days, but lately there was no need because non-volitile ram and
EEPROM's could hold data for hundreds of different home entertainment
devices.

Lately, there are new models coming out and the manufacturers can't
keep up...........so, there seems to be a resurgence in the
manufacture of 'learning' remotes.

I looked on ebay and there are many models that advertise that they
'learn' the code from the actual remote control to be cloned. A few
sport backlit screens and USB connectors so they can be updated over
the internet! I checked out a few of these in detail, and found that
they all rely on the basic codes stored in eprom and you have to
select what code you need for your particular device. When you run
into a function that you can't use for whatever reason, you have extra
buttons that can be customized (based on the output of the original
remote).

I was not clear whether you could roll your own completely from
scratch, or whether you had to start from the basic codes stored in
the remotes eeprom.

But, clearly....this is the wave of the future. We intend to purchase
one of these units soon.

While this can't help the original ops question regarding how to
design one....perhaps he could purchase one and dissect it to get a
starting place and go forward from there with his own design.

Regards,

M

On Fri, 27 May 2005 14:33:10 +0200, "Thomas W" <thomas_w@op.pl> wrote:

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on
the
Web but didn't find all I need. I wonder especially how the device will
be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the
project
for Monday.

Thanks in advance,
Thomas W.

 

[Mark Zenier]

In article <d77488$1uq$1@news.onet.pl>, Thomas W <thomas_w@op.pl> wrote:

Hi All,
I need to design an universal IR remote controller for my university
project. It should work with most of the devices available on market and
learn the codes from the original remotes. I did pretty much research on the
Web but didn't find all I need. I wonder especially how the device will be
able to store the signal codes, if coding differ depending on producer?
Any ideas? Maybe someone's done a similar project and will be willing to
share his knowledge? I'm pretty desperate since I have to finish the project
for Monday.

Go find the first issues of Circuit Cellar Ink magazine, and the Byte
magazines for the previous few months. (1987-1988?).

<http://www.circuitcellar.com>

There was a project for a Universal Remote in the last few "Circuit
Cellar" columns by Steve Ciarcia in Byte, and then an article, by Ed
Nisley, describing the 8051 firmware design in a followup article
in one of the first "Circuit Cellar Ink" magazine.

Hint: They make IR receiver modules with operating frequencies from
20 kHz to 95 kHz, so it may be more complicated than you think.

Mark Zenier mzenier@eskimo.com Washington State resident