Author Topic: Stainless steel as an absorber of EMI  (Read 827 times)

0 Members and 1 Guest are viewing this topic.

Offline Simon

  • Global Moderator
  • *****
  • Posts: 15105
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Stainless steel as an absorber of EMI
« on: May 27, 2020, 09:45:16 am »
I have been planning to make a box out of steel so that the circuitry inside emitting EMI will be absorbed and not make it's way out despite the box not being "water tight". I'm thinking of stainless steel instead to avoid having to tin or plate the steel. But I cannot find any data on it's absorption ability compared to steel. Steel is obviously several orders of magnitude better than copper or aluminium. A2 stainless is a bit magnetic so while not being steel grade, would it be better than aluminium.
 

Offline jpanhalt

  • Frequent Contributor
  • **
  • Posts: 832
  • Country: us
Re: Stainless steel as an absorber of EMI
« Reply #1 on: May 27, 2020, 11:31:31 am »
Type A2 is also called 304  or 18/8.  It is not the hardest to machine, but it does work harden easily.  If you need to drill holes, be sure to keep your feed rate up.

You might want to consider 416 stainless (https://www.pennstainless.com/resources/product-information/stainless-grades/400-series/416-stainless-steel/).  Clearly magnetic (martensite) and easier to machine.  My "stainless" refrigerator seems to be made out of that or something similar.
 

Offline imo

  • Super Contributor
  • ***
  • Posts: 2637
  • Country: 00
Re: Stainless steel as an absorber of EMI
« Reply #2 on: May 27, 2020, 11:35:23 am »
There are tables for the "skin depth" for various materials. Take the lowest frequency of your EMI and look at the materials required..

https://en.wikipedia.org/wiki/Skin_effect

 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 15105
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Stainless steel as an absorber of EMI
« Reply #3 on: May 27, 2020, 12:03:20 pm »
skin depth is more about the frequency, I am looking at the materials as their reflective versus absorbent ratio.
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 15435
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Stainless steel as an absorber of EMI
« Reply #4 on: May 27, 2020, 12:11:36 pm »
All conductors (as differentiated from superinsulators and superconductors*) are lossy, so it doesn't really matter.  You're actually making things worse by choosing a lower conductivity, but kind of maybe not it depends.

What are you doing, what are you trying to do, and how?

If you have waves building up inside the enclosure, it may be worth adding some absorber material, to dampen resonances.

Inside a typical sized boxy metal enclosure, these will be cavity resonances at UHF to microwave frequencies, with especially strong coupling to charged nodes or current loops in certain unlucky locations, depending on frequency (mode).

If you have waves building up inside, and little absorption inside, the waves may end up finding their way out of otherwise seemingly inert gaps, slots, holes, whatever.  EMI tape, springs, etc., and mechanical redesign (perhaps including more overlapping seams or fasteners), are good mitigation.

This all seems rather unlikely for what you're probably working with.  Hence the question; what are you doing, etc.


Far more likely, your concerns are down to conducted EMI along cables, and the enclosure could be plastic (perhaps metallized plastic) for all that matters.


*Superconductors exhibit zero resistance, thus perfectly reflect incident energy; this is only true at DC, and not quite true at AC.  The best superconducting resonators are made from spun niobium, and exhibit a Q factor in the 10^7 range at 800MHz.  Superconductivity does not have infinite bandwidth; it must, necessarily, drop off at some point, and that drop will be accompanied by losses.  This is typically in the THz to far-IR band, which is why superconductors -- high-temp ceramics in particular, which are always black -- do not suddenly become impossibly shiny when they cross below the critical temperature!

Conversely, superinsulators... aren't really a thing, sort of.  There are topological insulators, which are unexpectedly well-insulating, except for surface states if applicable.  Plain old vacuum, absent any charge-carrying matter**, is about as good as we can get.  After all, we can see light from the furthest reaches of the universe, and that's a pretty long trip.  Anyway, if we had lossless insulators of non-vacuum permittivity, we could use alternating layers to make a dielectric mirror (at least for a given frequency range) of arbitrarily high reflection and zero loss, to the same end.

The more fundamental point being, there is symmetry, an equivalence, between zero and infinite conductivity.

**Note that this is still a bit exceptional, as the best vacuums we can produce, still have cosmic ray radiation zipping through them; collisions with walls produce showers of charge carriers, and the particles themselves are often charged (e.g. super high energy protons).  These cause conductivity to be strictly nonzero over the long term.  Such conductivity might be low enough that we are unable to observe any impact on, say, transmission of low frequency EM waves, over any scale of vacuum vessel we can construct on Earth.***

***And just to get really into it... We have known for quite some time, that interplanetary space is in fact full of radiation.  Solar wind is a plasma, true it's tenuous compared to most lab vacuums -- but over astronomical distance scales, it behaves as just any other charged gas does, trapping magnetic fields, carrying momentum, shocking into surrounding (interstellar) gas, etc.  (And so on up the scales, to galactic clusters and intergalactic media, which are even lower density still, but nonetheless nonzero.)  It's noteworthy that plasma has a cutoff frequency, above which it doesn't block EM radiation; for the Earth's ionosphere, this is ca. 30MHz, which is why shortwave communication can bounce around the planet, but VHF+ flies out into space.  For interplanetary media, this cutoff is some kHz, above which we have relatively unimpeded view of the universe, and below which it just looks like you're surrounded by whatever noise the solar wind contains (much like trying to observe sunlight on an overcast day, or view IR radiation inside an oven).

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: exe

Offline Simon

  • Global Moderator
  • *****
  • Posts: 15105
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Stainless steel as an absorber of EMI
« Reply #5 on: May 27, 2020, 12:26:01 pm »
I have a small actuator that is in a plastic box, being a little brushed motor it will emit a bit of ooh nasty but not much but enough to not pass military. The idea is if we make a box that absobs rather than reflects like aluminium we have less need for a perfect faraday cage. That's all.
 

Offline TheUnnamedNewbie

  • Super Contributor
  • ***
  • Posts: 1010
  • Country: 00
Re: Stainless steel as an absorber of EMI
« Reply #6 on: May 27, 2020, 01:03:44 pm »
Any metal will still reflect most energy, even with various conductivities. Covering in absorbing material, be it carbon-based or iron-based, will prove to be much more effective in suppressing those effects than going for a different metal box.
The best part about magic is when it stops being magic and becomes science instead

"There was no road, but the people walked on it, and the road came to be, and the people followed it, for the road took the path of least resistance"
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 15105
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Stainless steel as an absorber of EMI
« Reply #7 on: May 27, 2020, 01:07:22 pm »
The ratio quickly makes a difference though as it becomes an exponential effect if you consider how many times will this emitted signal bounce before it is reduced to a level that is no longer a concern should it escape. Every time it bounces it could align with a gap just the right way (wrong way).
 

Offline TheUnnamedNewbie

  • Super Contributor
  • ***
  • Posts: 1010
  • Country: 00
Re: Stainless steel as an absorber of EMI
« Reply #8 on: May 27, 2020, 01:16:10 pm »
The ratio quickly makes a difference though as it becomes an exponential effect if you consider how many times will this emitted signal bounce before it is reduced to a level that is no longer a concern should it escape. Every time it bounces it could align with a gap just the right way (wrong way).

When you deal with a cavity in the size that I imagine you are dealing with, it's not like a laser that is bouncing around in a box, but you are instead exciting a resonant mode. It is not a single wavefront bouncing around and then happening to align with some hole. If anything, using a lower-conductivity material will lower the Q and make the entire thing more wideband, so now instead of a narrow strong resonance, you have a wideband weak resonance.

The best part about magic is when it stops being magic and becomes science instead

"There was no road, but the people walked on it, and the road came to be, and the people followed it, for the road took the path of least resistance"
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 15105
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Stainless steel as an absorber of EMI
« Reply #9 on: May 27, 2020, 01:19:38 pm »
so the absorbent materials are nowhere near steel in terms of their properties? I have seen the nice sheets but their self adhesive glue only goes to -20C. If the material is not an issue why do people use µmetal for serious stuff?
 

Offline TheUnnamedNewbie

  • Super Contributor
  • ***
  • Posts: 1010
  • Country: 00
Re: Stainless steel as an absorber of EMI
« Reply #10 on: May 27, 2020, 01:26:26 pm »
In order to absorb power, you need to not reflect it first. Absorbers tend to do that by
a) not being a smooth surface that reflects but using lots of tiny spheres or similar that scatters the impinging field in random directions (many of them not being 'away' from the absorber).
b) being quite lossy, so as the energy keeps bouncing around, it loses a lot of its power.

Because the conductive elements in absorbers are far smaller than the wavelength (eg, tiny carbon fibers, iron dust, etc), you can often actually model them as a more meta-material-like thing, a dielectric with very high loss tangent. You can't have the full 'surface' of an electron cloud you need to really 'reflect' the wave. That is the difference.

If you want to go nuts, you can also look at stuff like artificial impedance surfaces, where you can cut/mill slots or fingers into a surface to achieve a similar effect. I've seen this done with waveguide components, and used it myself in antennas.
The best part about magic is when it stops being magic and becomes science instead

"There was no road, but the people walked on it, and the road came to be, and the people followed it, for the road took the path of least resistance"
 

Online bdunham7

  • Frequent Contributor
  • **
  • Posts: 816
  • Country: us
Re: Stainless steel as an absorber of EMI
« Reply #11 on: May 27, 2020, 02:11:38 pm »
I have a small actuator that is in a plastic box, being a little brushed motor it will emit a bit of ooh nasty but not much but enough to not pass military. The idea is if we make a box that absobs rather than reflects like aluminium we have less need for a perfect faraday cage. That's all.

What is the spectrum of the EMI that is emitted?  Are there certain bands that are more important than others?  EMI from arcing is hard to deal with--how big is the motor?

I think you're on the right path with SS.  I think a common material for this type of application--keeping EMI in--is type 301 SS, which is weakly magnetic.  I don't know of any specific graphs or charts with the frequency-dependent reflection and absorption specs, but these people might if you contact their 'shielding team'.

www.tech-etch.com
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people.
 

Offline Wallace Gasiewicz

  • Regular Contributor
  • *
  • Posts: 126
  • Country: us
Re: Stainless steel as an absorber of EMI
« Reply #12 on: May 27, 2020, 02:18:40 pm »
Are there not some common stainless that absorb RF? Like the cooking pans that work with inductance stoves?
Maybe these types of stainless will work at your frequency

Wally
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 15105
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Stainless steel as an absorber of EMI
« Reply #13 on: May 27, 2020, 03:11:28 pm »
Well I have no idea at the moment and my 100MHz scope is probably not going to be much help. it's only 35mA at stall so no not much needed at all.
 

Online T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 15435
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Stainless steel as an absorber of EMI
« Reply #14 on: May 27, 2020, 03:13:53 pm »
I don't think you will find a low enough conductivity among any metal (or metallic alloy) on the periodic table, plutonium included!  (And yes I'm being serious -- there is actually quite a bit of public knowledge on this famously novel and toxic element.  Its low conductivity is just one of many peculiar properties!)

Even among metalloids it's a challenge.  (Part of the challenge there being, you need a stable allotrope or compound that isn't simply an insulator as well.)  Graphite is pretty reasonably conductive, enough that it probably isn't great by itself either.  But it can be used in a lower density form, such as flake, thread, etc., which helps as the filling can be a resin, which is strengthened by the graphite.  (Of course, metals or metal oxides can be employed in a similar fashion, or organic semiconductors -- conductive plastics -- and etc.)

Which is... a roundabout way of saying, yes, absorbent materials exist and can be useful. :-DD

But again, it doesn't matter, it's going to be all about conducted emission.  These are not frequencies where geometric optics help you any.

Like, how many months have you been banging on this stupid fan?  In this time, you could've:
- Completed a university E&M course, or
- Learned an EM field simulator and tested the geometry, or
- Built it, taken it to a test lab and proven whether it works or not, or at what frequencies the issue remains
- ???

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Online bdunham7

  • Frequent Contributor
  • **
  • Posts: 816
  • Country: us
Re: Stainless steel as an absorber of EMI
« Reply #15 on: May 27, 2020, 03:33:47 pm »
Well I have no idea at the moment and my 100MHz scope is probably not going to be much help. it's only 35mA at stall so no not much needed at all.

Assuming that's in response to me, I don't know that your scope won't help, although obviously it isn't the best tool.  You still might use the FFT to spot any peaks in whatever the motor is giving off, at least in the BW it is capable of.  I suspect you may find that the spectrum is low enough that absorption isn't a practical solution.  Is the case of the motor all metal?  Is it snubbed with a capacitor right at the motor?
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit DMM is good enough for most people.
 

Offline Simon

  • Global Moderator
  • *****
  • Posts: 15105
  • Country: gb
  • Did that just blow up? No? might work after all !!
    • Simon's Electronics
Re: Stainless steel as an absorber of EMI
« Reply #16 on: May 27, 2020, 03:45:54 pm »

Like, how many months have you been banging on this stupid fan?  In this time, you could've:
- Completed a university E&M course, or
- Learned an EM field simulator and tested the geometry, or
- Built it, taken it to a test lab and proven whether it works or not, or at what frequencies the issue remains
- ???

Tim

No it's not that fan it's a little brushed actuator. I just worry that I end up in a situation where anything radiated jumps my filter and becomes conducted anyway. Making a screened compartment for the filter in an already screened box would just become a bit silly.

Remember I work for a company where it can't cost more than a fiver!
 

Offline JohnG

  • Frequent Contributor
  • **
  • Posts: 257
  • Country: us
Re: Stainless steel as an absorber of EMI
« Reply #17 on: May 27, 2020, 09:34:09 pm »
The carbon impregnated antistatic foam used to ship ICs makes a surprisingly good absorber. You want the black foam, and it should show some conductivity with an ohmmeter. You can line the inside of the box.

John
"Those who learn the lessons of history are doomed to know when they are repeating the mistakes of the past." Putt's Law of History
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 11454
  • Country: us
  • DavidH
Re: Stainless steel as an absorber of EMI
« Reply #18 on: May 27, 2020, 10:44:41 pm »
What conductive material is used will not help with seams which form slot antennas.
 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 1199
  • Country: ua
Re: Stainless steel as an absorber of EMI
« Reply #19 on: May 28, 2020, 10:27:21 am »
skin depth is more about the frequency, I am looking at the materials as their reflective versus absorbent ratio.

it depends on type of field which you want to shield: far field or near field. If you want to protect from near field it depends on exact type of field: magnetic or electric.

In some case steel is better than copper, for other cases copper is better than steel. So it depends on the case - how far is field source (relative to wave length) and what type of field you want to shield.

The same shield can work better for one type of field and worse for other type of field. The same material can work better for one wavelength and worse for other wavelength.

Also it depends on shield thickness. So, there is no simple answer on that question. It depends on what you want to achieve.

What type of emission you want to shield? Is it far field from transmitter placed on a long distance? Is it near field from local source? What type of field you want to filter - electric or magnetic? What emission wave length you're want to shield? All this should be taken into account...
« Last Edit: May 28, 2020, 11:27:00 am by radiolistener »
 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 1199
  • Country: ua
Re: Stainless steel as an absorber of EMI
« Reply #20 on: May 28, 2020, 01:32:29 pm »
here is my estimation for stainless steel vs copper shield of 1 cm thickness (updated).
E and H field attenuation is estimated for 0.1 meter distance from the source.

PW = shield attenuation for plane wave (shield is placed at far field region of the source, at distance > lambda / 2)
H = shield attenuation for magnetic wave
E = shield attenuation for electric wave

[attachimg=1]
« Last Edit: May 28, 2020, 03:59:58 pm by radiolistener »
 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 1199
  • Country: ua
Re: Stainless steel as an absorber of EMI
« Reply #21 on: May 28, 2020, 04:43:15 pm »
I have a small actuator that is in a plastic box, being a little brushed motor it will emit a bit of ooh nasty but not much but enough to not pass military. The idea is if we make a box that absobs rather than reflects like aluminium we have less need for a perfect faraday cage. That's all.

shield attenuation depends on three components:

Shield Efficiency in dB = A(dB) + R(dB) + B(dB)

1) Absorption loss:

A(dB) = -20*log10( exp( t / δ ) )

where
t - shield thickness (in meters)
δ - skin depth (in meters)

δ = 1 / sqrt( pi * F * μ * σ )

where
F - frequency (in Hz)
μ - shield permeability
σ - shield conductivity

For most non-ferromagnetic substances (such as wood, plastic, glass, bone, copper aluminium, air and water) have a permeability almost equal to μ0.

2) Reflection loss:

R(dB) = -20 * log10( (1+k)^2 / (4 * k) )

where
k - impedance ratio

k = Zshield / Zw

Zshield = sqrt( 2 * pi * F * μ / σ )

For plane wave (far field region):
Zw = Z0 = sqrt( μ0 / ε0 ) = 376.73

For H wave can be approximated as:
Zw ~ 2 * pi * F * μ0 * r

For E wave can be approximated as:
Zw ~ 1 / sqrt( 2 * pi * F * ε0 * r )

where
μ0 - environment permeability, for vacuum μ0 = 4 * pi * 1e-7
ε0 - environment permittivity, for vacuum ε0 = 1 / (μ0 * c^2)
r - distance between shield and source
c - speed of light constant, c = 299792458

3) Multiple reflection correction term:

B(dB) = 20 * log10( 1 - exp( -2 * t / δ ) * (k-1)^2 / (k+1)^2
« Last Edit: May 28, 2020, 04:48:32 pm by radiolistener »
 

Offline radiolistener

  • Super Contributor
  • ***
  • Posts: 1199
  • Country: ua
Re: Stainless steel as an absorber of EMI
« Reply #22 on: May 28, 2020, 05:18:10 pm »
So, you can calculate absorption efficiency by using shield material permeability μ and conductivity σ, shield thickness t and frequency F:

A(dB) = -20*log10( exp( t * sqrt( pi * F * μ * σ ) ) )

If you put data from wiki: https://en.wikipedia.org/wiki/Electrical_resistivity_and_conductivity

For 1000 Hz and 1 mm shield thickness you will get:

Silver:
σ = 6.30E+07 [S/m]
A = -4.33 [dB]

Copper:
σ = 5.96E+07 [S/m]
A = -4.21 [dB]

Aluminium:
σ = 3.77E+07 [S/m]
A = -3.35 [dB]

Tin:
σ = 9.17E+06 [S/m]
A = -1.65 [dB]

Stainless steel
σ = 1.45E+06 [S/m]
A = -0.66 [dB]

The best absorption efficiency -4.33 dB is for silver, the worst absorption efficiency -0.66 dB is for stainless steel :)

You can use silver to get the best absorption efficiency in the shield.
But note that shield thickness also significantly affects absorption efficiency.
« Last Edit: May 28, 2020, 05:34:01 pm by radiolistener »
 

Offline imo

  • Super Contributor
  • ***
  • Posts: 2637
  • Country: 00
Re: Stainless steel as an absorber of EMI
« Reply #23 on: May 28, 2020, 07:19:04 pm »
During my diploma work I had to build a chamber for low signal measurements. A "soft iron" (I cannot remember the params) ~30mm thick plates were used, the box was around 30x30x25cm large. There was a single 5mm hole drilled through for the coax. As a proof of concept I put a small radio inside and we were listening through the hole. We were listening the music fine  :o I had to put another layer of copper shielding inside the box to silence the radio..  :D
 

Offline charliedelta

  • Contributor
  • Posts: 49
  • Country: ca
Re: Stainless steel as an absorber of EMI
« Reply #24 on: June 07, 2020, 10:48:28 am »
From one of my school books.
 
The following users thanked this post: macaba


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf