bad radio interference from computer

[flybike]

Hi,

One of my UAVs has the components from a mini desktop computer stuffed in it. It seems like overkill but I need it to run the software for a thermal camera. It is a microATX sized motherboard, AMD processor, and a little solid state hard drive.

The power source is the PST-ITX-3 PW-200-V from http://www.powerstream.com/mini-itx.htm running off a 3s lipo.

The problem is that it creates terrible radio noise and my 72 mhz radio becomes useless at anything more than about 100 meters. How can I reduce the interference? My first thought was to shield the computer in some sort of aluminum foil box, but I really know nothing about this subject. Would anyone care to chime in?

Thank you,
hans

[Unterhausen]

Did you make a UAV out of Pink Floyd?

Are you running the receiver off of the same battery as the computer? What motherboard are you using?

[flybike]

No, this is a 12 foot version of Pink Floyd. The receiver is on a different battery. The motherboard is http://www.newegg.com/product/product.asp?item=N82E16813135214

hans

[Unterhausen]

I guess that means it's putting out radiated noise that's saturating the input of your receiver. So you're going to put it in a box or wrap with tinfoil/etc. You probably want to minimize the number of wires coming out of the box, and make sure the cable to the imager is well connected. Have you tried it with and without the camera?

There are some people on here that know more about this than me, this is the Mechanical Engineering view of radio interference.

[flybike]

Without the camera it does the same thing. When the camera is turned on and the computer is not it has no interference.

[poynting]

It's a switching supply, so discounting noise from the processor/memory bus, I would expect most of your problems to be with the switching supply. Creating a faraday cage (wrapping the box with copper tape/tinfoil) should help with noise radiated from the computer side of the power supply, but it is likely that there is noise radiating out of the input side of the supply too. Go to radio shack and get some of these http://www.radioshack.com/search/index.jsp?kwCatId=&kw=rf%20choke&origkw=rf%20choke&sr=1 to put on the power lines from the battery to the switcher, as close to (or right at) the place the power lines go into the faraday cage you just made. You can also try adding some capacitance to the switcher side of the choke once it's installed.

If, however, the noise is coupled onto the camera lines coming out of the box, this won't help. In that case you're going to have a bigger problem, but chokes could help there too depending on the frequency you need the camera lines to pass.

Good luck!

[Unterhausen]

I admit that I'm not keeping up on switcher technology, but I would expect the rf coming from the switcher to be somewhere in the 20KHz-200KHz range and not much higher. I wouldn't expect that to interfere with 72MHz radio, but how can you be sure? So that's why I recommended putting it in the Faraday cage, if practical.

It seems like the computer is going to be putting out a lot of RF right in the range that includes 72MHz. PCI should put out 33MHz and harmonics, just as one example.

[zik]

Switchers put out frequencies not just on their switching frequency but on harmonics all the way up. Ham radio guys have a lot of problems with them and usually stick to non-switching power supplies as a result.

As poynting mentioned interference can also be emitted from any wires you have protruding from a device, even if the device is shielded. This can be hard to fix but the standard method is to use ferrites. If you're really keen you may want to read this paper: http://www.elmac.co.uk/pdfs/ferrite.pdf

[Unterhausen]

If I understand his system right, you could stuff the computer, lipo and ps inside a ball of aluminum foil with just the cable for the camera sticking out. The cable for the camera probably should have ferrites on it.

[flybike]

Yeah, that's what I was thinking. The camera cord is an ethernet cord. I have to do some reading on ferrites. How critical is it to have the aluminum foil box sealed? Not in terms of wires poking through, but just random little holes in it at the corners & stuff.

thanks,
hans

[Unterhausen]

random holes of a certain size are ok. Look at the size of your antenna, it's a fraction of a wavelength (1/4?) at 72MHz. You probably need some cooling holes, or you will cook the computer. Metal screen would be good for that.

[flybike]

If I were to switch to 2.4 ghz radio stuff would that be any better at dealing with interference?

[fly_asmara]

Hi flybike,

Before you switch to 2.4GHz, just curious, the 72MHz radio, this is your RC radio correct?

[flybike]

an update, maybe this will be useful for someone in the future-

I tried rf chokes on the two cables leaving the computer box and I wrapped the box in aluminum foil. I also placed the receiver 8 feet away from the computer. It all definitely helped but we still had range issues. Then I bought a 2.4 ghz RC radio system and tried it, it works perfectly even without any of the foil/chokes. $200 well spent.

hans