DIY Infra-red ? - Page 13 - RC Groups
Shop our Airplanes Products Drone Products Sales
Thread Tools
Mar 17, 2003, 03:50 AM
rebmeM roineS
MrBungle's Avatar

I haven't been around for a couple of months, so I have only now seen your post about programming your UV eraseable 12c509 chip, and the OSCCAL problem you were having.

Before you erase the chip, MAKE A NOTE OF THE OSCCAL VALUE!
Write it down on a peice of paper and glue it to the underside of the chip.

If/When you erase the chip, you will also erase the osccal value and it will need to be replaced at address 0x03FF! (0x01FF for '5x8 chips)

When I bought my UV-eraseable chips, and before writing any code to them, I did a hex dump(read) of the chips and saved them as hex files.
The saved hex file will have 0x0FFF in every address except at address 0x03FF, which has the 6 bit osccal value (for example, my 12ce518 had 0x0C88 at address 0x01FF)

After erasing my chips, I then write the OSCCAL hex file that I saved, and then the program code hex file.

Alternatively,(tho I've never tried it) you could also work out its value and write that directly to the OSCCAL register at the start of code. The OSCCAL value at address 0x03FF always starts with 0x0C, (because the opcode for MOVLW is 0x0Ckk) which leaves the lower byte as the value.
MOVLW 0x00kk and then a MOVWF 05 at the start of code should do the trick.

My CE518 was 0x0C88, the lowest byte is 88 which is 136 decimal.
Actualy, the osccal register only uses the higher 6 bits of the byte (the lower two are read as 0's) so the real value is 34 decimal.
Writing 34 to the osccal register will not work tho, must have something to fill in those lower two bits.

Sign up now
to remove ads between posts
Mar 21, 2003, 09:29 PM
Registered User
Koichi Tanaka's Avatar


As Jochen already tried, the simultaneous flight also according to an IR system yesterday was successful.

A gym has the area for three volleyball coats.

Even if TX were far and it was near, the airplane was able to be flown satisfactory.

Two planes carrying the 56.9KHz sensor and the 38KHz sensor performed the simultaneous flight.

PNA4612M(38KHz) and PNA4614M(56.9KHz) sensor was used.

56.9kHz TX only exchanges the 38kHz conventional PIC.

The 56.9kHz HEX file for 12C509A can be downloaded from my homepage.

A picture is a shot when controlling to each other in near.
Mar 23, 2003, 02:59 AM
Registered User
jberg's Avatar

great, I was already wondering wether anybody has noticed tis additional possibility. Now I would like to try flying on 33, 40 and 56kHz simultaneously. 33 and 40kHz are much closer than 38 and 56kHz, but I think it is worth a try. I just have to order the parts.


thanks for your offer regarding the video webspace, but I did not manage yet to prepare the file. I will come back to that.

Regards, Jochen
Mar 31, 2003, 08:08 PM
Registered User
Koichi Tanaka's Avatar
The HEX file of the 3 channel IR receiver corresponding to each maker's transmitter has been prepared.
It can download from my homepage.
Last edited by Koichi Tanaka; Mar 31, 2003 at 08:12 PM.
Apr 05, 2003, 08:16 AM
Registered User
jberg's Avatar

are you using one or two sensors in your receiver? It seems that you are using only one. I am using two, placed back to back and looking to the right and left. If you are using only one, does it look downwards? Do you experience glitches?

Regards, Jochen
Apr 05, 2003, 03:29 PM
Was that my 15 minutes?
Andy Birkett's Avatar
Hi Jochen,

I thought I'd chip in on the one sensor versus two issue. I think it all comes down to where you are flying and to some extent where you are standing!.

I'm not sure how big the regular Aeronutz venu is but if you fly with one sensor then you need to stand in the middle (lengthways) of the hall otherwise you can only use about 75% of the space. Most of the smaller (<13 gram planes) only have one sensor and even some of the bigger ones now use just one sensor. I believe in all these cases the sensor is pointing down.

With a two sensor arrangement it is 'traditional' to have then pointing left/right and inclined slightly down.

I think a lot depends on the (IR) reflective properties of the hall.

One sensor planes do appear to be more prone to 'dead spots' rather than glitches, often these are under a particlar light source, maybe where a bulb has been recently replaced?

I think if you see glitches (rather that areas of no response) then check the decoupling of the sensor from the main supply, some data sheets recommend 100R in series with 4.7F across the sensor.

Best Regards,

Apr 05, 2003, 08:53 PM
Registered User
Koichi Tanaka's Avatar

I think that explanation of Andy is right.

I set one IR sensor downward and have flown the airplane.

The lens of a sensor is shaved and the surface is made into the flat. It can be wide range by doing so, and a signal can be received now.

It flew satisfactory as well as the simultaneous flight also in the gym of the width of three volleyball coats.

Regard, Koichi
Apr 07, 2003, 06:18 PM
Registered User
Could you discribe your method of shaving the lens? Also does the detector window need to be polished after the lens is removed?
I wonder if doing the same thing to the emitters
would help also.
My current TX uses 48 emitters giving good coverage with two detectors back to back
mounted right and left.
I'm currently expermenting with a linear
Fresnel lens placed in front of the emitters
although not ment for IR i'm curious if it will
help outside at night.
If anyone writes the TX code for 40khz I would
like to try it also I notice quite a bit of dead band on the 3 channel receiver on the elevator
and rudder channels if someone rewrites the code maybe this can be reduced.
Thanks again to Koichi,Andy,Jochen and others
that have kept this thread going.

Regards, Joe
Apr 09, 2003, 09:45 PM
Registered User
I have some C code for the PIC to send and recive modulated IR in fact you can comucate with HP calculators with it.
Apr 10, 2003, 09:48 PM
Registered User
Koichi Tanaka's Avatar

IR sensor is made so that the infrared light from the surface may become the maximum sensitivity.
Therefore, a lens is on the surface of a sensor.

Sensitivity will become half if a PNA46xx sensor shifts from the front 37 degrees. Sensitivity will become zero if it shifts 60 degrees.

Can't IR sensor receive infrared rays broadly? I thought that it was that the conclusion removes a lens.

When the receiving range of infrared light was tested by IR sensor and the normal IR sensor which shaved the lens, the receiving range of IR sensor which shaved the lens clearly was wide.

The surface of the sensor shaved for the flat at first was polished.
I do not polish and think that it does not interfere.
The sensor of the bottom of a picture is not polished.

Regards, Koichi
Apr 12, 2003, 07:32 AM
Registered User
jberg's Avatar

thanks for the information and the pictures regarding your changes made to the IR sensors. I have wondered how you managed to make your receiver so light since one sensor alone weighs 0.4g, and if you use two of them that are already 0.8g. Now I am able to make my rx a lot lighter again.

Regards, Jochen
Apr 12, 2003, 11:04 AM
Sticky Shepherd
Graham Stabler's Avatar
surely a super small smt IR sensor must be in the works somewhere, phones and cameras are getting very small these days. We can but hope.

Apr 18, 2003, 01:26 AM
Registered User
Koichi Tanaka's Avatar
IR control module which has 38kHz and 56.9kHz switched was made.
The HEX file for PIC12C509A can be downloaded from my homepage.
Apr 18, 2003, 05:01 PM
registered user
DNA's Avatar
SMD IR sensor and transmitter..
Apr 18, 2003, 07:28 PM
Registered User
jberg's Avatar

unfortunately this device is most probably not what we need. The Panasonic and Vishay senors contain, besides the receiving diode, an IC with sophisticated preamplifier, gain control and demodulation. Therefore it needs a supply voltage and has a logic level output. The sensor at Electric Goldmine contains most probably only an emitting LED and a receiving diode. I believe it is impossible to make a preamplifier and demodulator lighter than within these special IC senors.

Regards, Jochen