PIC Beginner - Infrared Receiver and IR Protocol

[komoe01]

I'm a beginner for PIC Basic Program. I want to control my PC Program with IR Remote Control. In my place, we can get various IR remote control but a few type of PIC : 16F84A, 16F628A, 16F648A, 16F877A and 18F4550 can available.

I use 16F84A and 4MHz Crystal, RS232 for IR Remote Control.

My first program is IR Remote Receiver with Sony IR Protocol. I choose a DVD Remote Control (Made in China). I tested my receiver to my PC via COM port. It did not work. So I choose universal remote control and assign SONY. Now It's ok.

Secondly, I write my program with RC-5 Protocol. This code is posted by Bruce (http://www.picbasic.co.uk/forum/archive/index.php/t-308.html) It's also ok for universal remote that assigned for Philips. But it don't work for that DVD Remote Control.

Then I tested simply that DVD Remote Control's IR signal with LED. LED is worked between 900ms and 920 ms. It may be Header of IR train. I searched data of all protocols (http://www.sbprojects.com/knowledge/ir/ir.htm). I thought that DVD remote use NEC protocol. I also found that most of China DVD player's remote control use NEC IR protocol in a manual of DVD Player.

I learn NEC IR Protocol. It's type is different for Sony Protocol. Sony use Pulse Width Modulation [PWM]. But NEC use Pulse Distance Modulation [PDM]. I think that PulsIn is use to measure the pulse width. Is it right? So what is use to measure the distance between pulse. Is it [PulsIn + PAUSEUS]?

For NEC Protocol, I refer dhouston's post (http://www.picbasic.co.uk/forum/showthread.php?p=77859#post77859). My receiver's LED is ok for that DVD remote any key press but in my PC program all signal is Zero.

Any help would be greatly appreciated.

If this receiver is ok, I want to do Universal Infrared Receiver (UIR) with USB (with ref: eg. A universal algorithm for implementing an infrared decoder: www.edn.com/article/CA6653644.html.

Regard

Moe

[dhouston]

A schematic of your IR LED circuitry is necessary so we can see whether you need to measure pulses or spaces. PulsIn measures spaces as well as pulses - you merely need to indicate the polarirty. There are some example circuits here...http://davehouston.org/learn.htm

[komoe01]

Thanks for reply

IR LED circuitry is here.( please see attachment ).

I use this circuit and test Sony Protocol. LED is blink for any key and correct key code is accept in PC via RS232.

Here is code for Sony.

'start code for Sony Protocol --------------------------------------------------------------

'PIC 16F84A Sony SIRC Protocol

Include "Modedefs.Bas"
Define OSC 4 ' Set Xtal Frequency

Header var word
Body var word
i var byte
tmp var Byte
irButton var byte
irDevice var byte

irIN var PortA.2 'IR Input
ToCOM var PortB.4 'Serial Data to compute COM port [RS232]

'next time use both Sony and NEC Protocol in a PIC
'Same Key from different remote do different tasks
'eg Key 1 from Sony Remote ----> task1 in PC
' Key 1 from NEC Remote -----> task2 in PC

'SonyLED is blink for Sony and nec LED is blink for NEC Remote
SonyLED var PortB.2 'Green LED for Sony Remote keypress
necLED var PortB.3 'Red LED for NEC Remote keypress (Reserved)

TrisA = %00000100 ' Input RA2 only
TrisB = 0 ' Output All Port B

Start:
LOW SONYLED : LOW NECLED
IRBUTTON=255: IRDEVICE=255

'in Header is Sony 240
Pulsin irin,0,header ' Measure the header length.
if header > 270 or header < 200 then goto Start

'if correct header read body 12 bit
for i =0 to 11
pulsin irin,0,tmp ' Measure the body length.
if tmp >= 90 then
Body.0[i]=1 'in Sony Logic 1 is 120
else
Body.0[i]=0 'in Sony Logic 0 is 60
endif
next

IRBUTTON = Body & %01111111 'Mask 7 bit
IRDEVICE = (Body >>7) & %00011111 'Mask 5 bit

If IRDEVICE=255 then goto start

Serout ToCom ,N9600,[IRBUTTON] 'to PC COM - Port

if IRBUTTON >=0 and IRBUTTON < 255 then
high Sonyled
pause 300
low Sonyled
endif

pause 100
goto start


'end of code for Sony Protocol ------------------------------------------------------

I use this circuit again for NEC or CHINA DVD Remote Control.

' start code for NEC OR that DVD REMOTE -----------------------------
'PIC 16F84A NEC Protocol Protocol

Include "Modedefs.Bas"
Define OSC 4 ' Set Xtal Frequency

DEFINE PULSIN_MAX 968 '>968 RETURNS 0

Header var word
Body var word
i var byte
tmp var Byte
irButton var byte
irDevice var byte

irIN var PortA.2 'IR Input
ToCOM var PortB.4 'Serial Data to compute COM port [RS232]

'nec LED is blink for NEC Remote
necLED var PortB.3 'Red LED for NEC Remote keypress

TrisA = %00000100 ' Input RA2 only
TrisB = 0 ' Output All Port B

Start:
LOW NECLED
IRBUTTON=255: IRDEVICE=255

'in Header is NEC 900
Pulsin irin,0,header ' Measure the header length.
if header < 792 then goto Start

While irIN=0:Wend 'wait pulse

'if correct header read body 32 bit
for i =0 to 31
pulsin irin,0,tmp ' Measure the body length.

If ((tmp<40) Or (tmp>180)) Then start

if tmp >= 75 then
Body.0[i]=1
endif
next

IRBUTTON = Body & %11111111 'Mask 8 bit
IRDEVICE = (Body >>16) & %11111111 'Mask 8 bit

If IRDEVICE=255 then goto start

Serout ToCom ,N9600,[IRBUTTON] 'to PC COM - Port

if IRBUTTON >=0 and IRBUTTON < 255 then
high necLED
pause 300
low necLED
endif

pause 100
goto start

' end of code for NEC OR that DVD REMOTE ----------------------------

LED is blink for Both remote control. But Data in my PC software All keys are same number.

How can I fix it.

Thanks.

Moe

[dhouston]

From your schematic it appears you are using an IR receiver. Which one?

IR receivers usually are active low while the code I supplied for the NEC protocol was for RF where the receiver is active high. Also, my code was meant to work with the RF transmitter code that accompanied it which used slightly different timing than the NEC IR protocol.

The EDN article is wrong about details of the NEC IR protocol. It is explained in this NEC datasheet.http://davehouston.org/Nec6121.pdf
You will need to modify my RF code in order to receive/decode IR codes.
'12F629

@ __config _INTRC_OSC_NOCLKOUT & _WDT_OFF & _MCLRE_OFF & _CP_OFF

DEFINE PULSIN_MAX 1140 '>1140 RETURNS 0
DEFINE DEBUG_REG GPIO
DEFINE DEBUG_BIT 2 'GPIO.2
DEFINE DEBUG_MODE 1 'Inverted
DEFINE DEBUG_BAUD 9600
DEFINE OSCCAL_1K 1

IR VAR byte[4]
pulse VAR byte
i VAR byte
stx VAR word 'start of transmission

CMCON = 7 'comparators off
Debug "IR NEC PROTOCOL"
init: IR[0]=0:IR[1]=0:IR[2]=0:IR[3]=0:i=0
PulsIn GPIO.1, 0, stx
If (stx<760) Then init
'debug #stx
While GPIO.1=1:Wend 'wait space
Repeat
PulsIn GPIO.1, 1, pulse
If (pulse>100) Then
IR.0(i)=1 'set bit
EndIf
i=i+1
Until (i>31)
For i = 0 To 3
Debug IHEX2(IR[i] REV 8)
Next
GoTo init

EndI have not tested the above code (I don't have time to build a circuit with an IR receiver.) and you will need to convert it for your PIC and the pins you are using.

Note also that my code does not deal with the NEC shortcut repeat code.

[dhouston]

Change...
For i = 0 To 3
Debug IHEX2(IR[i] REV 8)
Nextto...
For i = 0 To 3
Debug IHEX2 IR[i]
NextThe bit order is already reversed as the bits are set by...
RF.0(i)=1

[lester]

Moe, you might want to read this article: Controlling the world from your armchair (http://www.picbasic.co.uk/support/Article.pdf)

[lester]

Moe, Infrared remote control decoding maybe considered something of a black art, however, this tutorial will show you that its principals are quite straightforward, and easy to implement on a PIC microcontroler.

Thanks to BASIC’s shallow learning curve, software designs that used to take weeks can now be realised in a just few hours. This article presents software solutions in the PicBASIC Pro language.

You might want to read the article: Controlling the world from your armchair (http://www.picbasic.co.uk/support/Article.pdf)

[lester]

The article is an extract from the book, EXPERIMENTING with the PICBASIC PRO COMPILER, written by Les Johnson.

The book is cuerrently out of publication, however we are considering publishing an online version as part of the new PIC BASIC portal that we are developing to support PIC BASIC Users.

[komoe01]

dhouston,
When I change code, old result 0 turn to 255. All the keys press are same.

http://img171.imageshack.us/img171/4220/pulsin.jpg


Header Logic 0 Logic 1 Logic 1 logic 0

2400 600 600 600 1200 600 1200 600 600
x a b c d e f g h i j
+------------------------+ +------+ +--------------+ +--------------+ +------+
| | | | | | | | | |
| | | | | | | | | |
| | | | | | | | | |
---+ +------+ +------+ +------+ +------+ +------
x____________________________|_____________|______ _____________|___________________|_____________|
pulsin header 1st loop 2nd loop 3rd loop 4th loop

return width of length





-- code start -------------------------------


pulsin irIn , 0 , Header 'Line 1

for i =0 to 3
pulsin irin, 0 , tmp 'line 3

if tmp > 90 then
body.0[i] = 1 'line 5
else
body.0[i] = 0 'line 7
endif
next

-- end of code -------------------------------

I think pulsin work as follow:

1. pulsin is timeout when no rising edge was found. return width is zero.
2. pulsin is time out when rising edge was found and then until falling edge. return the width length.


so pulsin in code run like that:


Line 1 : pulsin start measure (x) to end edge (b). Width is (a) to (b) so Header = 2400 (space (x) to (a) is not count.

Loop 0 (1st loop)
Line 3 : pulsin start measure (b) to end edge (d). Width is (c) to (d) so tmp = 600 (space (b) to (c) is not count.
Line 5 and 7: see the width lenght (not space) and decided logic 0
Loop 1 (2nd loop)
Line 3 : pulsin start measure (d) to end edge (f). Width is (e) to (f) so tmp = 1200 (space (d) to (e) is not count.
Line 5 and 7: see the width lenght (not space) and decide logic 1
Loop 2 (3rd loop)
Line 3 : pulsin start measure (f) to end edge (h). Width is (g) to (h) so tmp = 1200 (space (f) to (g) is not count.
Line 5 and 7: see the width lenght (not space) and decide logic 1
Loop 3 (4th loop)
Line 3 : pulsin start measure (h) to end edge (j). Width is (i) to (j) so tmp = 600 (space (h) to (i) is not count.
Line 5 and 7: see the width lenght (not space) and decide logic 0

PULSIN can only measure the width length.

When this code is used for NEC Protocol, all pulse are equal depending upon if statement (Logic 0 or 1)
Because In NEC, logic 1 or 0 is determined by space length. How can I measure space with pulsin.

lester
Thanks.

I already read Les Johnson's article and "Section 6: Experimenting with Remote Control" of that book. I already tested it. All are ok. But it's Sony IR Protocol (Pulse Width Modulation) that quite different from NEC IR Protocol (Pulse Distance Modulation).
I admire you and dhouston when I found your join date.
I start and learn picbasic at last week of July of this year. I search and collect PIC books (picbasic, C and Assembly) and article. I found about 60 books and articles. But I cannot find enough explanation for a command. eg. I want to know pulsin in detail. How pulsin work? examples relating pulsin. Pulsin in "PicBasic Pro Manual" is general. Picbasic is best language for me although I've little experience with C Language.


Moe

[komoe01]

Sorry sample wave is broken.

http://img171.imageshack.us/img171/4220/pulsin.jpg

If wave picture is not found, Please see it in attachment picture

Moe

[dhouston]

When I change code, old result 0 turn to 255. All the keys press are same.
It works fine for me. It's impossible for me to guess at what changes you have made to it. Try to test it using my code exactly as written. Once that works, try modifying it for your application, going step by step to isolate where the problem is being introduced.

The waves you've attached (including the second one) are not NEC IR protocol.

How are you sending the IR codes?

[komoe01]

dhouston

The wave image is a sample (part of the sony protocol).



Try to test it using my code exactly as written.

Ok. dhouston

But 12F629 pic is not available for me. Just a few PIC model can buy in my country. I have to use 16F84A or 16F628A. Beside I use IR, not RF.
I'm a beginner for pic so I've not familiar with DEBUG that used in your code.

So I write code to PIC than I see the result of LED signal and Computer program.

When I test it using your code exactly, what can I see the result from where? Can I use Infrared remote control? or I must built your transmitter to test receiver.

I think I must to change some code.

I do my best as you say!

Thanks.

Moe

[dhouston]

The code I posted above in this thread is for IR. If you study it you will see that the method is identical to Lester's.

I have one 12F629 sending the 32 bit codes and another receiving the codes. Instead of IR transmitter and receiver I have the transmitter PIC connected directly to the receiver PIC. I do not have an IR receiver handy. However, I've used the same basic code in devices which did have an IR receiver.

I quickly modified a simple test program written for other purposes to display the DEBUG output. See the attachment. You should see the same output ($50$AF$42$BD) every 15 secs) in whatever terminal program you are using.

As for DEBUG, read the manual - it's similar to SerOut and SerOut2.

You should be able to use my code merely by changing the GPIO.x pins to whatever RA.x pins you are using. I haven't used a 16F84A in ages so I'm not going to write the conversion for you.

[komoe01]

I changed pin assign.

The result is successfully done.

I test ten different kind of DVD Remote. Key positions are changed but all are work well.

Now I run my program for two different tasks using same key of two different remote (Sony and NEC Protocol).

** two ir protocol can use in same pic **

Adouston,
THANK YOU VERY MUCH for your kindly help.

I read NEC protocol and compare code.

Header is 9ms
- no header found, restart

If (stx<760) Then init

Space between header and body is 4.5ms
-wait space

While PortA.2=1:Wend

Each pulse is a 560µs long
A logical "1" takes 2.25ms to transmit (pulse 560µs + space )
a logical "0" is only half of that, being 1.125ms (pulse 560µs + space )

Please let me know:
1. How to calculate (pulse>100) is logic 1.
2. and PULSIN_MAX 1140 . why do you use 1140

I say again. Thanks.

Moe.

[dhouston]

Please let me know:
1. How to calculate (pulse>100) is logic 1.
2. and PULSIN_MAX 1140 . why do you use 1140I normally use RF where pulses & spaces are slightly different. I did the conversions to IR quickly, from memory and doing the math in my head so both figures are a bit on the high side.

IR transmitters (i.e. remotes) that used the NEC chip actually used numbers of carrier cycles as the unit of time rather than absolute number of µS. They were also battery operated so there could be some variation due to voltage variations. It is recommended to allow ±20% although that's probably generous with more modern remotes (the protocol has been around for about 40 years).

120% of 900 = 1080 instead of 1140.

With this method we are only able to measure the spaces which can be 0.56mS or 1.125mS. We just want to determine whether it is one or the other. A figure about halfway between the two values should work so 85 might be better than the 100 I used.

The NEC protocol is used by many manufacturers and you will encounter some that do not adhere exactly to the values given in the NEC datasheet so you need to build in a little flexibility.

[komoe01]

Thanks
dhouston
for your explanation.

Thanks
lester
for your suggestion.

Now I should stop this thread.


Moe

[tr6coug]

Lester, is there somewhere we can download the include files from the book? I'd like to use the Sony IR files in a project. Thanks!


The article is an extract from the book, EXPERIMENTING with the PICBASIC PRO COMPILER, written by Les Johnson.

The book is cuerrently out of publication, however we are considering publishing an online version as part of the new PIC BASIC portal that we are developing to support PIC BASIC Users.

[amenoera]

dears ,

i know its a very long time , but please am stick on this , can you post a full working code for pic16f84a to decode nec protocol .
many many thanks .