Xenon, Strobe on your Rc aircraft?

[rocky79]

Would a Xenon, Strobe on an Rc aircraft causes radio glitches or interference?
the strobe is similar to the ones used on camera flash unit.Does anyone know?
thanks for looking

[plane_spotter]

Looking around the forum and various model clips, high brightness white leds seem a very good substitute for xenon. Obviously not as bright but more than adequate for the job. Some of the multi-sequence strobes I have seen really do look impressive.

Using Leds for lights can cause radio i/f, so many recommend keeping wires short and the use of ferrite rings. At a guess xenon could also cause i/f. Just do a search on the forums, I am sure I read about it somewhere.

Good luck in your ventures. Maybe just try and experiment yourself on the workbench!!!

[z-matrix]

You will probably lose connection between your aircraft and transmitter while your strobe is flashing ( few micro seconds )
There is very high đi/đt thousands of amps going through the xenon tube at above 300Volts, for about a micro second,
oh and don't forget the ignition, a few kV needed to ignite the tube, this makes a spike at the reciever line, electrostatic protection diodes neccesera at the inputs of all equipment on the plane, and better put into metal foil shielded box.
( aluminium food wrap should be enough )
it depends on the design if the switched mode power supply causes radio interferences. ( you have to try it )

[sigve]

Hi

I buildt a strobe ligth out of an old camera once, and i dident notice any interference. My strobe ligth used a seperate battery and the cuircut was in no way conected to the rest of the reciever system.

sigve

[rocky79]

Hi

I buildt a strobe ligth out of an old camera once, and i dident notice any interference. My strobe ligth used a seperate battery and the cuircut was in no way conected to the rest of the reciever system.

sigve

Hello Sigve,
Did you put any shielding around the circuit? if so could you share the design
Thanks

[Gary Warner]

I have a Xenon flasher here at work. I layed on open o-scope probe over the flasher (200mhz scope). At 2cm from the flash tube, I get a 20ms pulse at .2 volts. It spikes in 10us and then decays in 20ms. The decay ramp looks like the discharge curve of an RC circuit. The chopper power supply used to charge the circuit emitts .05mv of radiation, seen all the time at about 10us.

Don't know if this helps or hurts. I'd like to do some read-world testing as I want this Xenon flasher in a giant-scale Beechcraft 200B.

Gary
--

[rysium]

I buildt a strobe ligth out of an old camera once, and i dident notice any interference.

I was reading about interference but finally decided to try it myself. I had a couple units from disposable cameras. I picked the simple unit with LED charge indicator. This circuit was so simple that in the 300V power supplier it had only one transistor.
I placed the unit 5 inches away from the RX antenna of the worse receiver I ever had (GWS Pico). I flashed the strobe many times and I didn't get a single visible glich. This way I think it is go for the strob for my giant scale Wilga.

[ZAGNUT]

what goes on in a strobe is no different than what goes on in a common CDI ignition used on gas powered planes. separate power supplies and maybe some shielding should be all that's needed.


dave

[westfw]

separate power supplies and maybe some shielding should be all that's needed.
I think the potential difficulty is shielding the flash tube itself. Essentially
you're talking about a large "spark gap transmitter", and the transparent
aluminum to shield the RF while letting the light through is a bit hard to
come by...

[Gary Warner]

I think the flash in an Xenon tube does not discharge in the form of rapid succesions of arcs, like free air arking does. Once the gas is ionized to conduction, I think the discharge is steady until conduction stops. This may lend this type of flash to a lesser amount of RFI then originaly thought.

Just thinking out loud.

Gary
--

[Acetronics]

Hi, Gary

You're right ...there's one very short arc ( micro to milli seconds ) until the applied High voltage goes under the minimum .
this works quite like the electronics Diac used in old light dimmers.

Here we fly a twin Jetcat 120, 3m40 Airbus A320 without any problems ... there are 6 flash units aboard, but all driven through opto-couplers, converters far from the radio components and fed by their own battery ...

Here it is: http://perso.wanadoo.fr/mes-avions-rc/

Alain

[rysium]

I flashed the strobe many times and I didn't get a single visible glich.

The simplest flash circuit I ever found made of Kodak disposable camera and a little changes. I tested the flasher again this time switching the transmitter off.

When I had just GWS Pico Rx on and TX off I didn't notice any signals in the air (servos stayed quitet and ESC was shut down). Then I turned on the flasher keeping the lamp about 12 inches away from the Rx antenna. I noticed that with most of the blinks there where servos reaction (kind of noise). About every ten'th flashes servos moved a bit and ESC jumped.

So it is a confirmation, that flasher unit makes interference but further range tests are required to establish how strong are they compared to the Tx signal. I will take Hitec dual conversion Rx (simple PPM, as PCM will cover the glitches) something like 555 and do the standard range check (150 ft with Tx antenna collapsed) with the flasher next to the Rx antenna. This should give me idea if the unit can be used in the plane.

Below you can find the strobe light circuit. I did some reverse engineering of the original unit (power supply part) but I don't know what size of capacitor is used for C1 (surface mount element). I modified high voltage circuit by adding R4 (20k) replacing original C2 (from 80uF to 10uF) and adding the autotrigger unit made of SCR and Neon lamp.

Smaller capacitor makes the energy of the flash 8 times lower. It's still high visible but should save the xenon lamp from overheating and extend it's life.

By adding resistor R4 I slowed down the charge rate. This slowed down the flashing rate to about one flash per two seconds. At the same time it reduced the battery current peek from over 2.5A to 0.8A (average is now about 0.3A).

The cool thing about it is that the unit came almost free - just for the cost of new capacitor, one resistor and one SCR. The rest is in the free camera (including neon lamp). If the ranbge check passes I'm going to design a new circuit board for all therse elements (taken out of the original Kodak board). I think the entire unit should fit into wingtip of my Wilga including 700mAh NiMh battery and the power switch. This way I will not need to guide a power cable through the entire wing length.

[rocky79]

You might want to look at intersil photo flash, they have a chip just for that.

[rysium]

Here is a deal. I found Luxeon 5W LED :eek: They are white and powerfull. I mean the flash from this LED has the same light effect as disposable flash camera. How do I know? I still see "stars" when I close my eyes. :D

I designed a simple circuit that runs the Luxeon 5W LED of 3 LiPo (Luxeon LED runs on 7V and the circuit limits the current to 700mA regardless of battery voltage as long as it is above 10V). The flash last about 100ms and repeats every 1 - 1.5 sec. 5W Luxeon LED is expensive (over 15 dollars a piece) but it's worth every penny. See the picture of this lights. They are small, light reliable and make no RF noise at all. It is just so cool :)

[rocky79]

Here is a deal. I found Luxeon 5W LED :eek: They are white and powerfull. I mean the flash from this LED has the same light effect as disposable flash camera. How do I know? I still see "stars" when I close my eyes. :D

I designed a simple circuit that runs the Luxeon 5W LED of 3 LiPo (Luxeon LED runs on 7V and the circuit limits the current to 700mA regardless of battery voltage as long as it is above 10V). The flash last about 100ms and repeats every 1 - 1.5 sec. 5W Luxeon LED is expensive (over 15 dollars a piece) but it's worth every penny. See the picture of this lights. They are small, light reliable and make no RF noise at all. It is just so cool :)

Wow, that is bright. Do you have schematic for the board and how does your circuit works?

[rysium]

Do you have schematic for the board and how does your circuit works?

Sorry, I thought it was clear. The schematic is on the paper showed on the picture. It doesn't have values because they are not really critical except the current regulator.

So here is the schematic with all parts and values. Blinking LED is from Radioshack (they have the only one). The MOSFET is from DigiKey - can be any (power type) as long as it can handle 1A and 11V. 1.8 ohm is 1/4W but I didn't notice any sign of being warm. The circuit works like on the picture (no heat-sinks) cold.

I designed the printed board (single sided) so I could use surface mount elements (the only "standard" element is Radioshack blinking LED.

The whole trick in this design is in Blinking LED. This is the cheapest, smallest and simpliest pulse generator. While the LED is blinking it takes much more current than idle - enough so the voltage on the 470R resistor triggers the MOSFET. I know, I'm briliant :D :D :D

Instead of using resistor in series with Luxeon LED I used LM317 together with 1.8R resistor that makes a current regulator for 0.7A. This way regardless of the battery voltage (as long as it is above 10V) the Luxeon LED blinks with full power, but always within the safe current limit (the LED is expensive, it would be a shame to smoke it).

[rocky79]

Sorry, I thought it was clear. The schematic is on the paper showed on the picture. It doesn't have values because they are not really critical except the current regulator.

So here is the schematic with all parts and values. Blinking LED is from Radioshack (they have the only one). The MOSFET is from DigiKey - can be any (power type) as long as it can handle 1A and 11V. 1.8 ohm is 1/4W but I didn't notice any sign of being warm. The circuit works like on the picture (no heat-sinks) cold.

I designed the printed board (single sided) so I could use surface mount elements (the only "standard" element is Radioshack blinking LED.

The whole trick in this design is in Blinking LED. This is the cheapest, smallest and simpliest pulse generator. While the LED is blinking it takes much more current than idle - enough so the voltage on the 470R resistor triggers the MOSFET. I know, I'm briliant :D :D :D

Instead of using resistor in series with Luxeon LED I used LM317 together with 1.8R resistor that makes a current regulator for 0.7A. This way regardless of the battery voltage (as long as it is above 10V) the Luxeon LED blinks with full power, but always within the safe current limit (the LED is expensive, it would be a shame to smoke it).

Interesting,I just don't understand the adjustable regulator configuration.How did you get the 0.7A through the LED?
Thanks

[rysium]

Interesting,I just don't understand the adjustable regulator configuration.How did you get the 0.7A through the LED?
Thanks

This is standard setup with LM317. LM317 is a adjustable voltage regulator that keeps 1.25V between Out and Adj. That means it limits the current on Out to keep 1.25V at the resistor 1.8ohm (1.25V / 1.8ohm = 0.694A). The drawback is that regulatopr requires about 2V between In and Out to work properly and there is 1.25 V voltage drop on the resistor, so the whole regulator circuit takes about 3V. This is why I need at least 10V power to run 7V Luxeon LED.

The voltage supplied from 3 LiPo is from 12V (fullly charged) to 10V (fairly discharged) so the power disposed (heat) is:

At 12V the voltage drop is 12V (power) - 7V (Luxeon LED) - 1.25V (Resistor 1.8ohm) = 3.75V so at 0.7A the power is 3.75V * 0.694A = 2.6W

At 10V the voltage drop is 1.75V, so the power is 1.75V * 0.694A = 1.21W

The power disposet at the resistor 1.8ohm is 1.25V * 0694A = 0.87W

It seems a lot, but remember the LED's blinking time is about 10% or less, so the total power disposed is less than 350mW.

The efficiecy is not the greatest:

At 12V it is 7V/12V = 0.58%
At 10V it is 7V/10V = 0.70%

but because of the high efficiency of Luxeon LED it is still great compared to simple bulb or even the xenon tube.

[Steph280]

That is a very interesting project. I am heading over to RadioShack later on.

One of the guy at the field actually took a true Xenon strobe from a life vest he had sitting around and mounted on his plane. It comes in a little watertight container similar to a flashlight, and uses 2 AA cells. So he took all the electronics out and attached a 2 cell AA battery holder. No mods needed.

[rocky79]

This is standard setup with LM317. LM317 is a adjustable voltage regulator that keeps 1.25V between Out and Adj. That means it limits the current on Out to keep 1.25V at the resistor 1.8ohm (1.25V / 1.8ohm = 0.694A). The drawback is that regulatopr requires about 2V between In and Out to work properly and there is 1.25 V voltage drop on the resistor, so the whole regulator circuit takes about 3V. This is why I need at least 10V power to run 7V Luxeon LED.

The voltage supplied from 3 LiPo is from 12V (fullly charged) to 10V (fairly discharged) so the power disposed (heat) is:

At 12V the voltage drop is 12V (power) - 7V (Luxeon LED) - 1.25V (Resistor 1.8ohm) = 3.75V so at 0.7A the power is 3.75V * 0.694A = 2.6W

At 10V the voltage drop is 1.75V, so the power is 1.75V * 0.694A = 1.21W

The power disposet at the resistor 1.8ohm is 1.25V * 0694A = 0.87W

It seems a lot, but remember the LED's blinking time is about 10% or less, so the total power disposed is less than 350mW.

The efficiecy is not the greatest:

At 12V it is 7V/12V = 0.58%
At 10V it is 7V/10V = 0.70%

but because of the high efficiency of Luxeon LED it is still great compared to simple bulb or even the xenon tube.

Great, Thanks for the explanation

[barb wire]

the IC is a TPS65552A . the Kodak camera capacitor in the circuit above (c/o rysium ) will discharge totally once the SCR turns on.
this new IC can allow multiple bursts of
the xenon flash tube. (this is how 'red-eye' is prevented )
instead of a SCR the flash circuit uses an IGBT (rated at 400V and
150 amps ) that can be triggered on and then off again before all the
photo flash capacitor energy is drained.

the texas instruments doc SLVA197 is a design tutorial on strobe
circuit design with this chip. turns ratios of the transformer, charging time, the charging diode, size of the output capacitor and PCB layout are some of the topics it covers.

one problem tho is the chips size. 10-MSOP
tough to solder without one of those hot air work stations
it can be soldered tho. using fine wires it can be attached to
a 'header' chip-socket thing with 0.1" pins and then the header
used as tho it were a DIP package

aries makes an 8-MSOP adapter to DIP 0.1" pin spacing but not a 10-MSOP
so you could solder 8 legs to it and air-wire the other 2, make that 1 as
pin 10 is a NC