17 replies to this topic

[headhunter23]

Yet another idea popped in my head a couple days ago. I remember watching some science channel ... I think TLC. The episode was how radio waves are affected by the size of a mesh covering over a fm or am radio(been a year or two since seeing it). I'm sure most of ya understand how it worked, guy tried different sized metal meshes and found one that was the same wave length or something that canceled all radio channels to the radio player. Ok, here's the idea, is it possible to just have a correct size mesh covering a receiver that would only allow then needed mhz or ghz through? I'm sure this idea has holes , but from what I understand less noise would mean better range and consistency.

I humbly await the reason why this wouldn't work. I'm talking to you mr rc-cam and ron and any other guru...

Edited by headhunter23, 17 October 2008 - 06:38 AM.

[Mr.RC-Cam]

It would be best to Google Faraday cage or Faraday shield. That will explain the principles of what the fellow had done.

[headhunter23]

So pretty much blocks everything, except when the holes in the mesh(dependent on size of holes) allows higher frequency waves through. So this could cut down on noise bellow a certain frequency correct depending on the mesh design? From what I've been reading it could cut quite a bit of noise. Am I reading this wrong?

Ivan.

[Mr.RC-Cam]

Traditional RF filters on the antenna input and elsewhere would be a typical solution.

What exactly are you trying to do? That is, do you have a specific noise issue you are trying to tame?

[headhunter23]

Nope, just had read that you guys did some testing as to the reason why leaving patch on ground increased range versus having patch on stand. From what I understood the measurments came out to be more noise reduction related than an increase in signal quality. Anyways long story short it seemed that decreasing noise had a big affect on range. So if noise on the receiver side in the airplane could be reduced then I would hope the range would also be increased by doing so. At this point just an idea... what are your thoughts on using a cage around a receiver providing the mesh allowed for the radio signal to get through? My thoughts are that the higher frequency's like the 2.4ghz would have a better time getting through the mesh, depending on the size(which I have no idea). But also I'm sure there would be negatives such as angles that the radio has to shoot through the mesh to hit the antenna might not always be prime, optimus prime that is...

Ivan.

[Doofer]

Well, in the example you give, the problem is that the 'noise' is Wifi, bluetooth and all that sort rubbish - at 2.4GHz. So no frequency filtering is going to help.

Keeping the aerials low puts the bulk of planet earth between the aerials and the extraneous groundbased sources, leaving the only clear path the one between the aerial and the 'plane. The cost, in my tests at least, is in absolute sensitivity i.e. aerials near the ground get less signal overal, but clearly proportionally less noise.

Not sure a Faraday cage would help further.

[Kilrah]

So if noise on the receiver side in the airplane could be reduced then I would hope the range would also be increased by doing so. At this point just an idea... what are your thoughts on using a cage around a receiver providing the mesh allowed for the radio signal to get through?

So, you mean the plane's R/C receiver?
Well, you already answered your own question earlier... A mesh will block the waves that are longer than the size of the mesh. If you design a mesh that will let the 35/40/72MHz of the R/C go through it, it will also let everything that has a shorter wavelength (= higher frequency) through. The 900MHz/2.4GHz of the nearby video TX will not be affected by it. Your complicated, heavy and large (it should include the RX antenna as well if you want to reduce total noise!) shield will only protect you from what is below R/C frequency (a few MHz), while not doing anything to the thousands of MHz that are above.
You're better off with the front end filters the receiver already has. What can be good is to shield the receiver's circuit itself, to prevent perturbations from acting on the RX circuitry after the front end filters, directly on the PCB traces. That's why many receivers use a metal shielding. It's rarely the case in R/C certainly due to weight/cost, but I've already seen some that had their case made out of a conductive plastic to somewhat act as a cage.

[Terry]

The idea of putting the patch on the ground to increase range was that it would reduce interference on the same frequency but not reduce the wanted signal. The cage would not help in this as it would reduce or not both the wanted signal and the interference signal the same.

Keep the idea's comming though, you never know


Terry

[headhunter23]

ahhh well will keep trying.

[Wavess]

...That's why many receivers use a metal shielding. It's rarely the case in R/C certainly due to weight/cost, but I've already seen some that had their case made out of a conductive plastic to somewhat act as a cage.

Well that´s fairly easy to do and not so heavy. Did you know that Magnetic Resonance use the Faraday cage for their rooms? Why do I say this? they prepare the room with copper everywhere with thin copper sheets and on the joints they use an adhesive copper "scotch type". I have one here and it´s very light so to cover a receiver with this tape it´s doable in weight terms. This copper light enclosure could be easy grounded to act as a faraday cage.

Another thought, why couldn´t we replace the heavy ferrite core the type 43 most used for an axial inductor with the same nH value? in this case it would be 850nH a very low value to match the mA required, but we should place at least to axial inductors right? one on the signal wire and the other at least on the V+ line.

BTW anybody knows which resonant frecuency works for the 43 ferrite core? For about the same value on the axial inductors we´re about 160-190mhz

[Mr.RC-Cam]

How did you decide on a 850nH inductor value?

[Wavess]

How did you decide on a 850nH inductor value?

Looking at the product specs here says 850nH

http://es.rs-online....B...&R=467-4166

but opening the pdf frin fair rite looks like 965nH, what I still don´t see is the resonant frecuency for the product

[Kilrah]

The AL value you state isn't an inductance in itself, but allows you to find the inductance of a coil made by wrapping N turns of wire around the said core.
This number in this case is most likely expressed in nH/N^2, they omitted the complete unit as is (too?) commonly done in that kind of very technical description, as engineers using it are supposed to know what they're talking about

BTW anybody knows which resonant frecuency works for the 43 ferrite core?

The "43 material" is one thing, but every differently shaped core made out of it will have a different AL value (and other characteristics).

Edited by Kilrah, 10 December 2008 - 04:28 AM.

[Mr.RC-Cam]

but opening the pdf frin fair rite looks like 965nH, what I still don´t see is the resonant frecuency for the product

I've never seen an inductor spec that listed it's operating resonant frequency since that is determined by its use in a circuit. However, the manufacturer's specification will normally list the inductor's self-resonant frequency (SRF). The SRF is an important characteristic that tells us the frequency where inductor becomes ineffective and is no longer a good inductor. To allow for best performance in a circuit, the designer would choose an inductor with a SRF that was twice the operating frequency. There are exceptions to this, but that is the general recommendation.

BTW anybody knows which resonant frecuency works for the 43 ferrite core?

Fair-Rite Type 43 material is formulated for broadband noise from about 30Mhz to 300Mhz or so. It is often a good pick for the types of issues we see in our application and I have used it to solve a variety of problems involving EMI/RFI noise. I found it was even a good pick for help in solving R/C Rx interference, as well as the GPS reception problem that 900Mhz users are experiencing.

[Wavess]

The AL value you state isn't an inductance in itself, but allows you to find the inductance of a coil made by wrapping N turns of wire around the said core.
This number in this case is most likely expressed in nH/N^2, they omitted the complete unit as is (too?) commonly done in that kind of very technical description, as engineers using it are supposed to know what they're talking about

Yes you´re correct on the AL factor it´s not the inductance itself but the equation it´s incorrect.
L=AL*N^2 where the inductance value would be in nH, so for my case I have 7 turns on my ferrite core it would be:
L=820*7^2=40.180nH=40,18uH
So if I try to select the same value on an axial stile inductor one, the Self Resonant Frecuency for 40uH would be 6.3mhz

I've never seen an inductor spec that listed it's operating resonant frequency since that is determined by its use in a circuit. However, the manufacturer's specification will normally list the inductor's self-resonant frequency (SRF). The SRF is an important characteristic that tells us the frequency where inductor becomes ineffective and is no longer a good inductor. To allow for best performance in a circuit, the designer would choose an inductor with a SRF that was twice the operating frequency. There are exceptions to this, but that is the general recommendation.


Fair-Rite Type 43 material is formulated for broadband noise from about 30Mhz to 300Mhz or so. It is often a good pick for the types of issues we see in our application and I have used it to solve a variety of problems involving EMI/RFI noise. I found it was even a good pick for help in solving R/C Rx interference, as well as the GPS reception problem that 900Mhz users are experiencing.

Yes on my example above we see that with the ferrite core linked above I would have about 40uH that would be aprox 6.3mhz SRF on an axial type of inductor, so theorically this would be effective to use in a circuit that works with 3.15mhz aprox?
Doesn´t have a relationship with our 35mhz, what I´m trying to get an aprox value to install some lighter and smaller inductors instead of the toroids one. If founded would it be correct to install them on signal and possitve?

[Mr.RC-Cam]

I would have about 40uH that would be aprox 6.3mhz SRF on an axial type of inductor, so theorically this would be effective to use in a circuit that works with 3.15mhz aprox?

That would be the usual recommendation. That is, keep the expected operating frequency under 3Mhz for that inductor. It's just a rule of thumb and a designer can decide what is best.

Doesn´t have a relationship with our 35mhz, what I´m trying to get an aprox value to install some lighter and smaller inductors instead of the toroids one. If founded would it be correct to install them on signal and possitve?

When a toroid is used I recommend that all the wires be included. That is because it is the expected configuration to reduce the common-mode noise. When discrete inductors are used then you could evaluate their effect as you install them on each conductor. Don't be surprised if you find you need them on all of them. But, every situation is different, so there is no way to predict what the needs will be.

In regards to the exact inductor value, just don't become married to a particular value. For reducing R/C glitches, a popular value is about 10uH. But I have seen cases where lower or higher values were better. It is just as important to carefully choose the inductor's other parameters; You'll want good current handling, SRF, and Q values.

For sure, what works for one person may not help another person out. Even if the two installations are "the same." You'll have to trust me on this one, since I have experienced the hardships that can occur when fixing EMI/RFI. Quite often the exact fix for one system is not quite the same for an identical system.

In a nutshell, there is no single silver bullet that can be used in every EMI/RFI situation. However, the methods used are the same. Getting the noise issues solved requires careful investigation and usually many controlled experiments. Plus an box full of Toroids and other filter trickery; when times get tough then out comes the copper tape, EMI gaskets, grounding straps, and prayers.

[Wavess]

In regards to the exact inductor value, just don't become married to a particular value. For reducing R/C glitches, a popular value is about 10uH. But I have seen cases where lower or higher values were better. It is just as important to carefully choose the inductor's other parameters; You'll want good current handling, SRF, and Q values.

Yes I have a bunch of inductors here to try with so

For sure, what works for one person may not help another person out. Even if the two installations are "the same." You'll have to trust me on this one, since I have experienced the hardships that can occur when fixing EMI/RFI. Quite often the exact fix for one system is not quite the same for an identical system.

In a nutshell, there is no single silver bullet that can be used in every EMI/RFI situation. However, the methods used are the same. Getting the noise issues solved requires careful investigation and usually many controlled experiments. Plus an box full of Toroids and other filter trickery; when times get tough then out comes the copper tape, EMI gaskets, grounding straps, and prayers.

I know and remember that we´ve talked about this many times. Trying to reduce noise onboard it´s hard to do, and requires many trial&error, the lasts trials cost me a new fpv gear due to an accidental short circuit . Also know that it doesn´t exist a single solution for our problems.
I´m preparing a small fpv plane now and I´m trying to get the more distance from the components as I can, so this makes some wires long, and probably some EMI/RFI would appear, due to my plane´s limited space and weight the toroids are not a nice solution for this case, so would have to try with axial inductors on wires.

[Mr.RC-Cam]

I´m preparing a small fpv plane now and I´m trying to get the more distance from the components as I can, so this makes some wires long, and probably some EMI/RFI would appear, due to my plane´s limited space and weight the toroids are not a nice solution for this case, so would have to try with axial inductors on wires.

Perhaps once you have it working noise-free you can post about the problems you ran into and how you solved them. The journey to tame the EMI/RFI demons is often interesting, so it would be grand to hear your story.