Is it possible to cover APM with aluminum foil to minimize UHF noise interference?

Is it possible to cover APM with aluminum foil to minimize UHF noise interference or is it bad idea because of its sensors, please?

I use Mini APM 2.7.2.

Regards

Views: 4487

Replies to This Discussion

I think it needs to be highlighted that this isn't just a squabble, pavel is talking complete nonsense and shouldn't be taken as something people need to do.

It is particularly unhelpful, for those who don't have the electronics skills to see it is uneccesary, to be given this bad advice.

The advice is generally especially usable for somebody who is getting only small control range or not reliable remote control with with open LRS remote control and some other UHF frequency band remote control systems. Also for sophisticated users which prefer improve things to maximum. The advice generally probably is not for user which likes plug and play working systems and which system is working in their usable remote control distance they usually fly.

Regarding if it is nonsense or not that is very questionable. Please read comments from Alasdair and R. D. Starwalt to find out their opinion also.

PS: I will probably try to do demonstrate some test which should hopefully prove if APM is able to cut remote control in some situations under some specific conditions or not when I will get a time. The practical answer should be yes or not.

Regards

No. If you come up with a crazy theory it is up to you to find some evidence BEFORE you claim that it is actually happening.

Or else it is a question and you can't argue with other people's experience that it does not occur.

I have restrained myself up to this point, I was waiting to see if you had a unique instance of a radiating APM, but from the generalised responses that you have posted can only deduce that you have a little knowledge of a big subject and unfortunately have decided that tinfoil will cure any actual or perceived RC problems...

... I have actually tested the HK Micro APM with a 'real' FR spectrum analyser (not the frequency sniffer capabilities of an OpenLRS module) and have been unable to detect any spurious RF on 390MHz - 490MHz (which should cover all OpenLRS frequencies that you may be using).

Hence I would repeat my original comment, with a slight correction... is your APM actually radiating on OpenLRS frequencies?

If the answer is no (as would appear to be the case based upon my own measurements), then the answer to your original question is No, it is a bad idea.

I'm sorry to say Ben but RF noise due to clock oscillators in digital equipment is basic electronic theory. There are really two general kinds of digital interference: that due to the "clock" oscillators in computers, peripherals and other digital devices, and that which is caused by the processing in a digital device. They are just another form of direct interference that is caused by leakage of radio signals at the digital clock frequency or some multiple and interference caused by the operation of the internal processor (or digital signal processor) in the device. The interference is not necessarily at a multiple of the clock frequency; instead it can be related to the rate at which instructions are being processed. The level and character of the interference can vary greatly with the type and amount of processing occurring, and the interference can disappear entirely if the digital device is not actively processing. Digital interference is usually relatively easy to identify. The signals involved are not strong and rarely affect systems at distances greater than about 6 feet.
Now the question as to wether this is applicable to us is uav design is a different matter. Suitable placement of components goes a long way in tackling the problem but shielding is also a valid point for discussion. For a more technical explanation please see Alasdairs post in this thread.

Pavel proposition is not crazy. There will be some interference, it's just too small to be an issue in most cases. The solution, is a partial solution, not perfect. The testing is ad-hoc, but I know plenty of Hams that use another radio to scan for interference, light bulbs as current limiters etc.. It's part of the 'hobby' side of figure stuff out.

The problem may be only be apparent with interference generating components close together, distance is the best in all cases as the interference drops of so quickly with weak signals.

The only problem with this thread is the commentary that moves from 'interesting' type discussion and critique, to playground bullying. That doesn't usually happen that much on this site.

As for tin foil hats, it you know anything about RF, you need a tin foil suit to be protected, maybe ;-p

No, the fact is that some electronics give off RF interference but there is no sign that the APM gives off any significant amount.

What makes Pavel crazy is that he has done no test to identify the APM as a source whatsoever and yet still claims it is despite wide experience and actual tests with the proper equipment show it is not one. He states other components which are known offenders such as the mobius/runcam, then falsely assumes that the flight controller must be too.

To ask the question is not stupid. To dispute people who have tested it and found it not to be the case when he has no evidence at all is very foolish.

I invested in RFExplorers when they initially came out and used them to troubleshoot a lot of EMI issues.  After spending a hundred hours scanning every piece of equipment in my fleet I narrowed it down to 2 particular components - MAX7456 chip on our beloved minimosds, and interestingly enough - anything running STM32 controllers at a 72mhz clockspeed.  In short, I detected harmonics on the 433 band any time either of those 2 components were anywhere near.

You needed about 5-6 inches of separation to get clear of them, but using something like aluminum tape (for heating ducts) did help to a tiny degree - oh and yes - I discovered later copper tape is better too :)

These were the results of my modifications...

https://www.youtube.com/watch?v=THwE4V0L6TI

Interestingly enough - I had multiple problems going on, the primary one being power outages to the uhf receiver - connecting it directly to the power rail fixed those power issues.   As for UHF control - that was entirely dependent on how far out you were and the proximity of the EMI emitting sources to the receiver.

Thought I'd just share that little tidbit.

Oh and for what it's worth - I never detected anything coming off any of my APMs :)  It was the supporting equipment that was the source.

Hi All,

In one of my previous jobs I actually had to work quite a lot with RF and EMF shielding and it is important to understand a few things.

Shielding can either absorb or reflect RF energy at selected wavelengths.

EMF and RF are not the same thing.

Aluminum can dampen EMF but is generally reflective to RF depending on frequency.

Steel is reflective of RF and interacts very strongly with EMF.

When either EMF or RF is absorbed it is converted to heat.

Graphite (or carbon) paint is one of the best absorbing means of shielding especially when applied to an electrically conductive grounded substrate (aluminum, steel, etc).

Mu Metal is a nearly perfect EMF absorber (you cannot put a magnetometer inside a Mu metal box because it cannot see the earths magnetic field).

The trouble with reflective shielding is that it is reflective and will reflect internally produced RF back on to the enclosed device itself.

Basically graphite, carbon paint over a conductive grounded shell is the best cheapest general purpose broad spectrum shielding material you can get.

It is what makes Stealth Aircraft work.

Best,

Gary

I would like to add my comments to the forum - It does seem that there is a lot of confusion and rumor surrounding the subject of EMC and poor Pavel is being needlessly hammered as a result..

I am not going to debate the APM's ability to interfere electromagnetically, or not, with other integrated systems - it is irrelevant for the discussion. The question was, will tin foil reduce radiated interference or not. The many comments and answers given need a little massaging and filtering..

Antennae, RF design and EMC testing formed the mainstay of my career for almost 30 years. If anyone wishes for more detailed info on my ‘credentials’ I will gladly oblige.

All digital electronics, APM included…, will generate a level of electromagnetic interference. It starts with a crystal oscillator, the basic clock for all the electronics. This will have a fundamental frequency ( indicated on the can) but is normally a square wave output. Fourier Functions dictate that this will have harmonics high up into the UHF/SHF region. The rest of the digital electronics , as the word digital implies, is of a square wave switching nature, at various sub and higher harmonics of the crystal fundamental frequency. In addition, the digital switching edges are very fast, from nanoseconds up. Again, Fourier dictates harmonics way into the SHF regions as a result. Since there is strong cyclic correlation between the crystal frequency and the digital electronics processing cycles, there will be higher levels of electromagnetic ‘noise’ at certain frequencies than others. However, as not all processes are synchronous, there will be lower levels over a very broad spectrum as well.  All this is true for any digital apparatus, autopilot, cameras, brushless motor controller, etc. The latter is of special concern, due to the fact that the currents being switched are so much higher, and so the generated EMI is higher.  The levels of interference are directly related to the energy input – A digital ‘switch’ switching many milliamps ( old TTL or Schottky Logic) will generate far higher EMI levels than a low power Cmos logic cct switching microamps.

Sound design techniques , PCB with ground planes top and bottom, low power logic, on-board LC filters on all I/O connections, etc, go a long way to ensuring the bare electronic module is as ‘quite’ as possible.

How does the electromagnetic ‘noise’ radiate from the electronic module? In all instances there are only two ways – Radiated and Conducted emissions.

Radiated emissions are identical to the RF emission from the antenna connected to your video TX and the like. If a conductor is attached to a source of RF,  the electromagnetic field results from the radiation by that conductor. Every copper track on the PCB is a potential radiator, as is every wire connected to the module – your servo wires, power cables, GPS cable, etc.

Conducted emission is interference conducted within a wire, remaining ‘internal’ to that wire, and entering an attached piece of equipment, and interfering with the normal operation of said equipment. Such interference is NOT shielded by means of a Faraday cage, but more easily by LC filters – Pi and L filters. Conducted emissions, at the risk of generalizing , are mostly related to power supply lines, and tend to be of lower frequency – such as within cables providing current to power supplies, etc.

If we wish to prevent interference from a piece of equipment from detrimentally affecting other systems by ‘shielding ‘ the culprit, there are a few basic rules to follow.

Place it in a Faraday cage. A Faraday cage is not some arbitrary metal box surrounding the noisemaker. Contrary to what has been said, a Faraday cage can consist of any metal, depending on what the aim is. A mild steel cage will shield against RF as well as pure magnetic fields , so is not advised for your magnetometer..An aluminium or copper shield will provide superior shielding against RF ( it is more conductive than steel) and no shielding against magnetic fields ( ok for your magnetometer). Aluminium is not a ‘poorer’ shield ‘because it reflects the energy and does not absorb it’…A faraday cage does not work by absorption – it is an RF sealed enclosure, and it contains all the radiated emissions within. The key here is that it is a SEALED enclosure. What does sealed mean? It does not have to be water or air-tight, but must be RF-Tight. That means that any gaps, holes or non contact points anywhere in the enclosure must be of a size at least 1/20 of the wavelength of the HIGHEST frequency you wish to shield at. So a wooden frame covered with chicken mesh of say 6cm grid would be fine for for shielding up into the lower VHF region, but not for 433MHz up.

Also, the Faraday cage does NOT need to be connected to 'ground', since the RF is contained within - there is no RF on the outside of a properly implemented Faraday cage.

If we assume an enclosure in a cube shape, with solid sheet copper sides, soldered along all edges, we then have a very good Faraday shield. However, with our electronics inside, although we have no radiated interference, we also have no external connection to the electronics. And this is where all the posted talk on the shielding by means of ‘wrapping’ falls apart.

In most cases the radiated interference is not from the module itself, but from the attached wires. If the module is placed within an effective Faraday shield, and a hole made in the side of the shield and the module wiring brought out through that hole, then all is lost…The wires will simply conduct the emissions outside of the cage and radiated externally . A Faraday shield ONLY works if all energy is contained within – this means extending the shield externally, around the wires, by means perhaps of co-axial or similar shielded cables, with the shield attached to the Faraday cage. However, for this to work, the other end of the cable/wiring must have its shield attached to the Faraday shield of the destination electronics…The Faraday shield only works if ALL sources of EMI are contained within, ALL along the way.

As has been said, shielding can absorb or reflect/contain electromagnetic energy. A solid plane shield of good conductivity ( copper, aluminium,..gold..) will reflect all energy at all wavelengths – it is not frequency sensitive. When a reflective shield is perforated , for example, to allow for air flow for air-cooling, or to save weight, the shield becomes frequency sensitive, and the openings must be around 1/20 at the highest attenuation frequency required.

Absorptive ‘shielding’ works by converting the impinging energy to heat, thus dissipating any further travel of the energy. This method tends to be highly frequency selective, due to the nature of the absorptive coating, its conductivity, is thickness, etc. A simple ‘coat’ of ‘carbon’ paint will do little, if the characteristics of the coating are not matched to the application, the depth of coating not controlled, etc.

Some Terms :

There is some confusion, it seems, related to the various acronyms connected with these subjects:

EMF – The proper definition is Electro- Motive Force, and has nothing to do with the subject at hand. EMF is the measure of Voltage between the two terminals of a voltage source. Unfortunately,  this acronym is being misused to mean Electro –Magnetic Force, and is as such quite incorrect.

Gary is quite correct when he says ‘EMF and RF are not the same thing’.

However, he then confuses the issue by using them interchangeably –

Aluminum can dampen EMF but is generally reflective to RF depending on frequency.

Nothing can ‘dampen’ EMF, since it is a measure of voltage potential..

Also, aluminium is always reflective to RF, unless it is used as a conductor of said RF. If used as the solid panels of a Faraday cage, it will always contain that energy, regardless of the frequency in use. Steel is a ‘worse’ reflector of RF than aluminium, but effective nonetheless, if clean, ie, not rusted, etc, Remember , any RF induced currents are carried in the very surface layer of the conductor – the skin effect, and the current depth is dependent on the frequency- the higher the frequency the thinner the skin depth. The poorer the conductor, ie, the higher the resistance of the material, the poorer the material will act as a reflector or conductor.

Steel does not interact strongly with EMF – there is nothing to interact with a measurement of Voltage..

EMF cannot be absorbed, so cannot be converted to heat – However, a load ( resistor) can be connected between the poles of a source of EMF, causing a current flow, and hence heat..

Also, stealth technology relies on two aspects – reflection, the greatest factor, and absorption. You will notice that there are few ‘flat’ on-angle surfaces on any stealth system – this ensures that any reflected energy is not in the same direction as the energy source. However, since it is not possible to have zero on-source surfaces ( such as the air intake nacelle edges)  these surfaces are kept as small as possible, and the absorptive coating takes care of the rest.

EMI – Electro Magnetic Interference. The measure of interference caused by Electro Magnetic waves on a subject equipment.

EMC – Electro Magnetic Compatibility. In order for various equipment to co-exist within a system, they need to comply with some requirements in terms of the interference each generates, and the level of interference each can tolerate. In the hobby world, it is a free for all – you would worry about it only if you experienced some level of problem. In the military and commercial environments Standards have evolved that allow manufactures of equipment to develop their equipment in isolation to other developers, knowing that if they comply with said standards, their equipment will neither cause nor be susceptible to outside interference from equipment likewise complying to the standards.

There are a number of such standards :

The American Mil Standards, such as  MIL-STD 461G ( soon to be F), primarily  a Military Compliance testing standard.

The Commercial Standards RTCA-DO-160G – used extensively in the commercial Aviation industry – All commercial Avionics has to be type certified to DO-160.

British Standards – BS EN 61000-4-1 / -4-3 / -4-6 – Similar to the MIL-STD-461, but far more pragmatic and applicable.

None of these standards have any bearing on our ‘hobby’ since the costs to comply are simply silly, but if you peruse them you will notice the importance of EMC, and the extensive use of Faraday cages – almost every piece of Avionics is housed within an Aluminium enclosure, forming a full Faraday cage, and all connectors exiting the ‘cage’ have signal pins filtered by means of L and Pi filters.

Since all our Hobby electronics is of open construction, these problems will plague…

I have meaured EMC of the early PixHawk, according to DO-160C, but only on the bench in my lab, with my HP-E7404A - 10GHz EMC analyzer.

According to DO-160C is stretching it a bit ( a lot..) since proper testing requires a proper EMC test chamber. However, if the tests are done comparatively, ie, measurements done with equipment off, and then with equipment on, on the results subtracted from the background noise, the results are meaningful and usable. I also live on the West Coast on Namiba, in a very small town, so is is electrically pretty quiet out here...

The PixHawk does not comply, by a large margin...this is with GPS attached, and wires attached to each Servo output, and to the serail I/O.

Does this mean the PixHawk is unusable?? Obviously not. It all depends where the emissions lie in frequency, and what the levels are. RF levels reduce by the inverse square to distance, and attenuation in air by the same law with increasing frequency, so spacing of equipment, levels of emissions, frequency, etc, all play a role in determining if a system installation is self-compliant or not. And each installation WILL be different, placement, wire routing, etc.

Sorry if there are some typo’s – I did try to get them all.

And sorry for the long story – RF is my passion..Maybe my story is not of much help, maybe it helps some understand the fundamentals a little.

 

The Nampilot…

Folks, 

Just about everyone contributing on this post is barking up the wrong tree!

The original post asked two questions:

1/. “Is it possible to cover APM with aluminum foil to minimize UHF noise interference”

2/. “is it bad idea because of its sensors”

Both I and a couple of other posters have succinctly responded to these questions.

For the former the answer must be NO, as the APM does not create any discernible interference at the frequencies Pavel is worried about, i.e. OpenLRS RC control operates between 430MHz and 460MHz.

For the latter, the answer would YES it is a bad idea, as encasing the APM in a metallic shroud could cause a number of unwanted issues, ranging from potential electrical shorts, interference with the correct operation of the barometric sensor and possibly the on-board compass.

All the discussion regarding faraday cages, attenuation of spurious electrical signals (however formed initially), and the like have done nothing to provide a clear and concise response to the original question.

Pavel has demonstrated his unfamiliarity with the APM (asking basic questions on the HobbyKing website regarding the APM), be honest does anyone really believe that the wealth of ‘knowledge*’ posted so far on this topic will help him?

He needed a clear yes, or no.

* (most of it I fully acknowledge is far beyond my limited experience or theoretical understanding)

@Kev

Problem is that many of the views expressed begin to establish themselves as fact which leads to further confusion.

The question was :

Is it possible to cover APM with aluminum foil to minimize UHF noise interference or is it bad idea because of its sensors, please?

It was not a question as to his assumption that the APM causing interference was correct or not.

Although Pavel might not be knowledgeable in the EMI field, he is smart enough to avoid the obvious, such a shorts by wrapping directly in foil...he indicated so very early on.

The Magnetometer would NOT experience any detrimental effects since Pavel wished to use aluminium foil...

The Airdata sensors would only have problems if the Faraday cage was airtight, which, as I indicated, it does not have to be.. So in fact, the correct answer all his questions is Yes!. It might not solve his problem, since maybe it is not the APM generating the interference he suffers from, but that was not the question..

Your intimation that the wealth of 'knowledge'  is wasted is unfounded - If views are expressed that are not quite correct, or maybe totally so, without further debate, then how do people progress and learn positively? I believe this entire posting was very beneficial - it showed clearly the lack of understanding and knowledge in this field, and how easily fiction becomes fact.

Many people have stated the APM does not cause interference - it would be nice to see some factual measurements, as some have stated has been done. One implementation , way of routing and connecting wiring, etc, may have problems due to its method, while others not.

It would be great if more assistance and mentoring took place, rather than the demeaning approach shown Pavel..

The Nampilot.

RSS

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service