DIY Infra-red ? - Page 4 - RC Groups
Thread Tools
Dec 30, 2002, 04:45 AM
Thats NOT indestructible
j8m8l's Avatar

Does that lil PCB include a Dc-Dc?

You could try and sand down the PCB so that it is a lil thinner (although you have probably already done that )


Last edited by j8m8l; Dec 30, 2002 at 06:01 AM.
Sign up now
to remove ads between posts
Dec 30, 2002, 06:06 AM
Registered User
jberg's Avatar

even thinner? The material is only 0.5mm thick!

Regards, Jochen
Dec 30, 2002, 06:12 AM
Thats NOT indestructible
j8m8l's Avatar
OK , not thinner LOL

I presume that it has the Dc-Dc, (it has the 3 Caps)


Dec 30, 2002, 06:35 AM
Registered User
jberg's Avatar
Yes, it has. At the lower edge in the picture, starting from the left side:

-output filter cap, 10uF multilayer ceramic
-LTC1754-5, 6-legged SOT23, 5V/50mA max. output
-pump cap, 1uF multilayer ceramic
-input filter cap, 10uF multilayer ceramic

Best possibility for weight reduction would be the elimination of the huge areas on the PCB that are not covered by parts, or using a double-sided board of half the size. And of course, using less solder (but I'm using already 0.5mm solder wire and a lighted magnifying glass; perhaps using solder paste and soldering all parts in one with a heat gun would help).

All that makes the building more difficult and time-consuming, but at the low price of all the parts (the whole rx is at about 14EUR) I want to build several, one for each model. This fits well to self-made actuators which are also cheap. So every model can be fitted with its own complete rx system, which is good because in my experience a model from which the rx system is taken out (for a new model) will never get one again.

Regards, Jochen
Last edited by jberg; Dec 30, 2002 at 06:43 AM.
Dec 30, 2002, 07:01 AM
Only nerd in the village
Nice freehand PCB drawing This is one advantage with SMDs - you can almost draw the PCB from the diagram. What do you etch with?

Is the 50mah from the dc/dc enough for driving the esc and two coils or will you use rudder only?

OK, I take my headband idea back. Would look nerdy though Have you given any thought to using remote transmitter arrays like seen on Tysons site? This would certainly increase range.

Everyone: Three cheers for Koichi, Jochen and Andy for sharing this information with us.

Dec 30, 2002, 07:03 AM
Thats NOT indestructible
j8m8l's Avatar


Oh and Jberg, thanks for answering all my stupid questions

(The about goes to michael as well!!)

Dec 30, 2002, 07:49 AM
Registered User
jberg's Avatar
Thanks, you're welcome!

Josh, there are no stupid questions, only stupid answers!

I'm standing on the shoulders of giants here! Andy and Koichi have done the hard parts. Now I'm really curious for an in-flight test ...


I have drawn the circuit with a special felt-tip pen (a "laquer marker" with a relatively thick, flowing ink) directly onto the bare copper. Using a foil and the photo method is better for multiple boards, but I had speed in mind yesterday. The hard part are the connections to the SOT23 parts (FET and DC booster) which I had to redo several times since the feet pitch is only 0.95mm.

I use iron-III-Cloride and the good old marmelade glass method: Put the unetched board into a small glass with screw-on cap (I actually used a glass that contained cloves of garlic in vinegar before - your taste may vary ). Add iron-III-cloride (a relatively concentrated solution in water) up to about half of the PCB (not more, or there will be less effect), close the glass and heat everything in hot water or with your heat gun until you are just still able to touch it. Then shake well for some minutes and you can see the copper going away very quickly and with sharp contours.

I always pour the solution back into my stock bottle (one liter of solution will work almost forever) - and be sure to give the solution to your harmful substance reception office once it doesn't work anymore - this is poisonous!

Andy, Koichi, once the whole system has proven to work we really should put everything together on one website!

Regards, Jochen
Dec 30, 2002, 08:18 AM
Only nerd in the village
I have made lots of PCBs in the past using a special pen like you describe. I gave it some thought and I think this might work for the SMD chip pads: Draw a full line to the total outline of the pads and then use a pointed object to remove the ink between the pads.

I have used a mix of hydrochloric acid and hydrogen peroxide to etch boards ultra fast (dangerous!) but I think this will be too agressive for these tiny PCBs. I will try your method.

Good idea about a webpage but who has the time for it? I think it would be great if someone would also sell the pre-programmed PIC chips.

Andy; how about some code that would provide normal servo outputs? The micro truck brigade would love it and I think there are still those in the ranks of indoor flyers who are a bit put off by actuators.


Too many ideas - too little time
Dec 30, 2002, 08:40 AM
Registered User
Originally posted by jberg
Andy, Koichi, once the whole system has proven to work we really should put everything together on one website!

Regards, Jochen
Sounds like great idea for those of us who are very interested in making something like this (with some help from a fellow expert of mine).

Dec 30, 2002, 08:39 PM
Registered User
The only problem using the pen method is...them pin spacings are getting mighty fine. I'd love to have a laser 'printer', like the guys are using to cut kits. I think you could 'etch' a whole batch of boards that way, and not worry about tiny parts. I concur on a 'servo' version...that way everyone will have no excuses for not using it.
Dec 31, 2002, 03:32 AM
Registered User
Koichi Tanaka's Avatar

CC5XFREE compiler The combination of MPLAB was successful.
Thank you very much.

Regards, Koichi
Dec 31, 2002, 06:10 AM
Was that my 15 minutes?
Andy Birkett's Avatar
Back home again so a chance to catch up!

I believe the purpose of the 38Khz modulation is two-fold:

First it allows the system to eliminate interference from mains lighting at 50/60Hz and provides some tolerance to sun light although strong sun light will swamp the receiver).

The other function is allows the transmitter LED to be pulsed with a high current. If by maths is correct the I/R transmitter is only on for 13µS at a time (and then off for 13µS) allowing several amps to be passed through it. I think my Ztron tx uses 15 I/R LED with around 2.7 amps going through them, they get warm in use. They are wired as 5 lots of "3 LEDs in series with a 2.2R resistor" ( 12V - 2V*3 => 6V across the 2.2R resistor => 2.7A?) When you measure with a meter you get the average current. Of coarse if you TX jams on then you blow your LEDs - check for blown LEDs by using a video camera it can see I/R (a digital camera may work?).

The different modulation frequencies would not allow multiple planes at once, you would need I/R transmitters and receivers at different wavelength of IR. I suspect just using different modulation frequencies would still cause too much interference since ultimately you are just flashing a lamp on and off the ons and offs would collide.

The receiver’s code is tolerant to some noise but will cut off completely if there is no valid signal for around 1 second. Both the ESC and actuators have more steps of control than the Ztron system but use a lower frequency.

Regarding programmers I've used the JDM hardware with the IC-PROG software for some time see the links at can make an adapter for the surface mount PICS using a PLCC socket, again see the above link.

Jochen, is the code modification the same as you made for the ACT Pico 4uP 35MHz receiver / Futaba FC-16 combination? I would like to combine the code all back into one source file and then post the hex files as well as the source so that it can be directly blown into the PIC and people don't need to bother with compilers etc.

When Graham surfaces from his vacation I think you also need to consider higher actuator drive currents. I use 130R coils directly from the PIC (they don't blow up honest) but the same coil (or even a 200R one) with a proper drive circuit is much more powerful.

Since this solution is practically identical to the radio decoder that we are working on for the GWS maybe it could all be combined into one solution???

Michael, servos would be much easier than actuators! I think we need to get Jochens modifications combined into the original source and then take it from there.

I've got about 200M of spare webspace if you want to dump it all there!


Dec 31, 2002, 07:13 AM
Registered User
jberg's Avatar

yes, the modifications are basically the same, although I have fiddled around some more with the maximum/center/minimum pulse widths to get it going with my older Multiplex Pico Tx that doesn't have programmable trim, so the pulse widths really have to fit.

Now that I know the correct values I really would like to get back as close as possible to your original solution with the auto-setup code. Probably I will do that directly after the first in-flight test of my rx .

I think I will program a PIC that just simulates a Tx signal at 33kHz and try how much my rx will be disturbed by that signal. I have a plug-in breadboard so it is quite easy to build up that system. It would be very nice and help a lot to have at least two IR systems working in the same room. carries some 56kHz IR sensors, maybe that helps.

Regards, Jochen
Last edited by jberg; Dec 31, 2002 at 07:16 AM.
Dec 31, 2002, 07:45 AM
Only nerd in the village
Originally posted by Andy Birkett

When Graham surfaces from his vacation I think you also need to consider higher actuator drive currents. I use 130R coils directly from the PIC (they don't blow up honest) but the same coil (or even a 200R one) with a proper drive circuit is much more powerful.
Zetex makes a nice H-bridge that should work well. I will get Graham to order some for me. Jochen suggested making a H-bridge from four Micro3 sized fets - order to Uncle Conrad on the way. More powerful actuators are good for ailerons but we don't want to spend all the power on moving surfaces, so a H-bridge from transistors might be sufficient (just like the one on the Bit RX).

Andy; I realise servo output is simpler - no need to do PWM.

I have no student TX interface in my TX. How do these things work? Would it be possible to hook up the IR TX to a TX without interface somehow?

Dec 31, 2002, 10:48 AM
Sticky Shepherd
Graham Stabler's Avatar
I am here lurking but my parents dial up connection frustrates me. I'll be in my new house at the weekend. To complicate matters further I have just discovered that the Aeronutz meet is this Saturday and I have nothing to fly and all my stuff is 2 hours drive away.