Apologies I've been off air!!

Excellent idea Koichi

Yes it should all work. I think all you are doing is modulating the signal using IR rather that rf.

In the case of radio the GWS takes the rf modulation out leaving a signal for the decoder to work on. With IR, the sensor does the same thing, removing the 38Khz compoment.

If Jochen's experiments yeild good results then for the existing Ztron receivers you would simply need to replace the PIC and job done.

Further more you should be able to get all 8 channels . You would need to piggy back a second PIC (and possible third) (with slightly modified code to decode the highr channel numbers).

By the way, if you are adapting existing ztron receivers, look out for the difference in PIC packaging. Most of the Ztron ones use wide (208mil) PIC the GWS decoder boards use smaller 150mil PICS.

Andy

FWIW I also think Andys code is superior to that of the original Ztron. I goes to full duty cycle on the actuator outputs making them a bit more powerful. I also belive the throttle is a but smoother?

This might just get me back on IR again (were it not for the fact that it still doesn't work in daylight) - I always disliked the fact that I could not use a "proper" transmitter with nice joysticks. Are the IR RX sensors still readily available? I read somewhere about Panasonic discontinuing some of their sensors but I don't know if they are the ones we use?

I still have a Ztron IR RX with 3xservo outputs so I would be interested in knowing if the TX would work with this.

Michael

Andy,

thanks for your comments. I'm already thinking about what can be done with my now lighter equipment.

Michael,

the IR emitters and sensors are matching items "Made in Germany" by Vishay-Telefunken, not Panasonic, and are available from Uncle Conrad. The datasheets tell that the IR emitters are high-performance ones and that the sensors are especially good in filtering out unwanted stuff, like daylight(!), tungsten light and fluorescent light. The transmitting distance with one(!) IR emitter is said to be up to 35m, so by using several of them we should be on the safe side. I will make some experiments on this and will keep you informed.

The weight of one sensor (most of the legs removed) is 0.2g. These sensors are available for a carrier frequency of 33, 36, 38, 40 and 56kHz, unfortunately the 56kHz type is not carried by Conrad, but perhaps this will even enable us to use more than one IR control system at a time! I will also make some experiments on this.

The system shown in my picture is amazing, you are not able to break the connection by covering the emitters or sensors by your hand or by pointing the emitters away from the sensor. I did not make transmission distance trials yet, though.

The Conrad (http://www.conrad.com) item codes:

-emitters: TSAL6200, #171140-62, 0.56EUR each (5 pcs or more)
-sensors: TSOP1838, #171115-62, 1.51EUR each

Yes, I also like to use my real Tx with all the bells and whistles (v-tail mixer and so on). The crystal can be removed, so the Tx will not send, no disturbance of other modelers.

I'll keep you informed.

Regards, Jochen

Hi Jberg

This is truly amazing

can you just explan somthing to me...

Is this how the system works?

1)The recived IR signal is converted into a standard signal as with a normal rf rx (pic one)

2)Pic 2 converts this into pulse for the motor and actuators (this is andys one?

I would love to see how this works out, if we could get like 4-5 frequencys it would solve all the problems with IR that we have now (well almost)

(I thinkt that i am going to try all of this some time in the future, should make a great A-Level DT project )

Hi,

I just tried the system outside. Using 4 emitters at a peak current of 120mA (one could go to 200mA without problems) the connection works at up to 35m *outside* at 3.00pm (right now). We have heavy clouds here at the moment, but no rain. When putting the rx offset to about 45-60° from the direction where the tx is the distance goes down to about the half. You have to loosely point the tx to the rx for these results. Inside a normal living room it doesn't matter where the rx and tx point to, the connection will always be up.

This is good news IMHO! I'm sure these results can be further improved by using more emitters, two or three sensors, and higher current. The tx uses a duty cycle of about 8%, so 2 sets of emitters driven by 200mA peak each will result in only 32mA constant current draw, which can easily be provided by the tx battery, also because the original rf tx is deactivated by pulling out the crystal.

Now I *have to* build a flyable rx!

Josh,

only one PIC is needed! The sensor contains all the electronics to make a nice digital signal from the IR signal. This signal is fed directly into one PIC that contains Andy's code (slightly modified by me). That's all!

Regards, Jochen

This is good news indeed! Does not matter if it will not work in bright sunlight - we don't get much of that around here anyway

Here is a serious suggestion (although it might sound like a joke): How about mounting an array of IR transmitters on a hat or a headband. You are more likely to be pointing your head at the plane than you are at pointing the TX in the right direction. This would also be a serious fashion statement; wearable R/C

Is the 200mah rating for the transmitters peak or continuous?

Michael

OK

Fantastic Jberg (sorry i can be a bit slow at catching on!!)

Why did Koichi have like 2 Pics? (or did he not )

Do you think that the TX can be used with all TX's or would the code have to be changed

Can anyone reccomend a PIC programmer that is cheap that can do smt pics?

Wearable is a good idea, it has been used for video RX's



http://www.nyblimp.com/atv/atv2.htm

Josh

Michael,

regarding the headband tx: You wouldn't be able to look sideways for a moment, but you are able to look sideways and point the tx to the model!

Direct sunlight is always a problem for micro models as sunlight generates wind, which in turn makes flying the really small stuff impossible. But being able to fly at sundawn when the wind has calmed down would be nice, and will probably work.

The emitters are rated 100mA continuous, 200mA at 50% duty cycle and up to 1.5A for single 100us pulses, but efficiency will go down above 200mA.

These are the datasheets:
http://www.vishay.com/document/81010/81010.pdf
http://www.vishay.com/document/82047/82047.pdf

Regards, Jochen

I would think it would work with any Tx, just change the cord.

As to a PIC programmer...there are plenty of them, just get one, and then you can get an adapter for socket->SMT. I'd think you could rubber band it in place to program it.

Originally posted by j8m8l
OK

Fantastic Jberg (sorry i can be a bit slow at catching on!!)

Why did Koichi have like 2 Pics? (or did he not )

Do you think that the TX can be used with all TX's or would the code have to be changed

Can anyone reccomend a PIC programmer that is cheap that can do smt pics?

Wearable is a good idea, it has been used for video RX's


Josh

Ok fellows, I just have sat down today and have made one.

The picture shows an "Andy Birkett" PIC, an ESC and a DC/DC booster, all on one small PCB that I have made today. All-up weight as shown is 1.0g plus the shown sensors (0.35g each) and all the necessary cables that are always too heavy. And yes, it works.

Regards, Jochen

The bottom left solder joint looks 'cold'.

are all these bits in one place? I see what could become a standard board, that needs code. Looking at Andy's site, looks like C code, which i would have to compile, then burn? I think it would be nice if there were one spot to go, where you could d/l a foil pattern, and a precompiled file, all ready to go.

Jochen,

Surprise is felt for your quick test.
It is impressed also by the wonderfulness of Andy's PIC program.
If I do not use two PIC, I cannot make.
The CC5XFREE compiler was downloaded.
File compiled by CC55XFREE can it use by MPLAB?
Please let me know the compile option of CC5XFREE in that case.

Lightweight IR receiver was built. It is 0.76g including a cable.
This IR receiver is also using two PIC.

Regard, Koichi

Originally posted by Mighty Mik
The bottom left solder joint looks 'cold'.

You have sharp eyes, but it is *not* cold. There I have soldered and de-soldered the IR sensor for testing, so the joint isn't as smooth as it was before.

Koichi,

Originally posted by Koichi Tanaka
Surprise is felt for your quick test.
It is impressed also by the wonderfulness of Andy's PIC program.
If I do not use two PIC, I cannot make.
The CC5XFREE compiler was downloaded.
File compiled by CC55XFREE can it use by MPLAB?
Please let me know the compile option of CC5XFREE in that case.

Lightweight IR receiver was built. ...

I was quick because you and Andy already have done the hard work! Thanks again!

Yes, the CC5X compiler does work integrated into MPLAB V5.4 and V5.7 (not with V6.1). Just follow the "INSTALL.TXT" file. I will send you the compiler options soon.

Please note that I have modified Andy's code slightly to fit my tx (this has nothing to do with the use of IR instead of RF).

The rx would be lighter without the PCB which weighs about 0.5g but it is easier to build with a PCB. I want to supply every model with its own receiver (which is possible because this rx is so cheap) so I will probably build several.

Regards, Jochen