Author Topic: Faraday cage, shielding and grounding  (Read 9744 times)

0 Members and 1 Guest are viewing this topic.

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Faraday cage, shielding and grounding
« on: June 09, 2020, 11:07:46 am »
Hi,
I need to protect a small experimental space so that it is immune to external noise as much as possible. I had few ideas which then raised a few questions:

1) what is the best material? Copper mesh, metal mesh, solid copper, solid steel, other?

2) is placing a Faraday cage inside another more effective? Or would it create some capacitor (or other) undesired effect?

3) if yes to (2) above should they be connected to each other? Of would connecting each other create some waves bouncing around inside between them and create other effects that ultimately affect the inner most space?

4) should I be grounding it or leave it floating? Or perhaps if (2) is a better solution should I ground perhaps the external one and leave floating the inner one?

Thank you :)
« Last Edit: August 13, 2020, 06:53:54 pm by ricko_uk »
 

Offline Henrik_V

  • Regular Contributor
  • *
  • Posts: 160
  • Country: de
  • “ground” is a convenient fantasy
Re: Faraday cage, shielding and grounding
« Reply #1 on: June 09, 2020, 12:37:02 pm »
Two cases, isolated by 10-20 mm styrofoam , outer to earth, inner to guard 

outer case 5mm Aluminium ... and good contact of the plates...   also will shield magnetic fields ...

Another construction used here: two boxes made of transformer iron 'foil' (~2mm) , like shoe box , isolation again 10-20 mm

If you place heat generating parts in there add forced ventilation ... (aluminium mesh..)   temperature monitoring is a good idea anyway
Greetings from Germany
Henrik

The number you have dialed is imaginary, please turn your phone 90° and dial again!
 

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Faraday cage, shielding and grounding
« Reply #2 on: June 09, 2020, 02:06:31 pm »
Thank you Henrik for your suggestions/solutions,

few questions:
1) why styrofoam and not just "empty space"?
2) puraly out of curiosity, why 10-20mm gap and not less or more? Just practical construction or is it related to some waves/physics phenomenon?
3) you mentioned "outer case 5mm Aluminium ... and good contact of the plates...   also will shield magnetic fields ..." I don't think aluminium blocks magnetic fields. Are you referring to some special aluminium alloy perhaps?

Many thanks :)
 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #3 on: June 09, 2020, 02:14:26 pm »
Two cases, isolated by 10-20 mm styrofoam , outer to earth, inner to guard 

outer case 5mm Aluminium ... and good contact of the plates...   also will shield magnetic fields ...

Another construction used here: two boxes made of transformer iron 'foil' (~2mm) , like shoe box , isolation again 10-20 mm

If you place heat generating parts in there add forced ventilation ... (aluminium mesh..)   temperature monitoring is a good idea anyway
Aluminum is bad because it forms an oxide. 

Copper is probably best.  You can solder the seams.  Place an FM radio inside the box and you can test for leaks.  I have used copper sheet or copper screen.  The screen is nice because you can see inside the enclosure.  It also provides some ventilation.

What frequencies are you trying to shield?

The  door needs to make a good connection all the way around.  Maybe multiple latches. 

Double shielding works better but is it worth it? 

Do you need to bring signals our power into your screen box?  That's always a weak point for shielding. 

 

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Faraday cage, shielding and grounding
« Reply #4 on: June 09, 2020, 02:53:51 pm »
Thank you Ahbushnell :)

What frequencies are you trying to shield?

As broadband as possible, from Hz to GHz (WifI etc). It needs to be as quiet as we can make it without spending thousands of dollars.

Double shielding works better but is it worth it? 

If it contributes then definitely yes, especially if the materials cost is only few tens of dollars.

Any feedback on Henrik suggestion about styrofoam and connecting?

Do you need to bring signals our power into your screen box?  That's always a weak point for shielding.

Yes, but an analog one not a power one (battery is inside the enclosure for that reason). I just posted another post exactly about this if you have any suggestions they are very much appreciates, you might want to have a look at it here: https://www.eevblog.com/forum/projects/optically-transparent-electrical-isolation-of-analog-signal/?topicseen

Thank you :)
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21231
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Faraday cage, shielding and grounding
« Reply #5 on: June 09, 2020, 03:16:35 pm »
Hi,
I need to protect a small experimental space so that it is immune to external noise as much as possible. I had few ideas which then raised a few questions:

1) what is the best material? Copper mesh, metal mesh, solid copper, solid steel, other?

2) is placing a Faraday cage inside another more effective? Or would it create some capacitor (or other) undesired effect?

3) if yes to (2) above should they be connected to each other? Of would connecting each other create some waves bouncing around inside between them and create other effects that ultimately affect the inner most space?

4) should I be grounding it or leave it floating? Or perhaps if (2) is a better solution should I ground perhaps the external one and leave floating the inner one?

Thank you :)

This is really an EMI/EMC issue. Many small details contribute to the overall efficacy of such enclosures.

It would probably help to read one or more of the standard textbooks on the topic. One that has been through several revisions over the decades is Ralph Morrison's "Grounding and Shielding Techniques in Instrumentation". That might help, and if not there are others.

It is at times like this that the demise of decent technical bookshops is felt. Once upon a time it would have been possible to go and skim a couple of books, then buy the one that looked most suitable.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: ricko_uk

Offline trevatxtal

  • Contributor
  • Posts: 34
  • Country: england
Re: Faraday cage, shielding and grounding
« Reply #6 on: June 09, 2020, 03:58:53 pm »
How about a unused microwave oven ?
Made safe of course.
 
The following users thanked this post: ricko_uk

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 4034
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #7 on: June 09, 2020, 04:21:31 pm »
Most microwave ovens are not great.  I've tested a cell phone inside a microwave and it still works.  They are made to be cheap and good enough for purpose not a lab grade screening cage.

Anyway for low frequency your limit will be magnetic shielding.  You will need a magnetic material like iron, steel, or mumetal to get good attenuation for Hz range magnet fields and it will be very hard to get good low frequency magnetic screening over a large volume.  Low frequency electric fields are easy to screen.  For high frequency the limiting factor will be your signal feedthroughs and the seams -- especially the door.  Real screenrooms usually have berillium copper  finger springs all the way around the door and compress them with a lot of force.  EMI gasket material is an easier alternative that can also be effective but may wear out with use and stop making good contact.  Springs are good because they scrape away oxide and contamination each time they are used to make sure the contact is good.
 
The following users thanked this post: ricko_uk

Offline fourfathom

  • Super Contributor
  • ***
  • Posts: 2005
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #8 on: June 09, 2020, 04:38:16 pm »
Most microwave ovens are not great.  I've tested a cell phone inside a microwave and it still works.  They are made to be cheap and good enough for purpose not a lab grade screening cage.

Microwave ovens are designed to contain the microwaves, not all random RF fields.  The door seals are usually a resonant gap that provides a virtual short-circuit at the specific microwave frequency used (2.45 GHz).  While this gap works very well at this frequency, it will not provide much if any attenuation at other frequencies.

So it's not a matter of cheap, it's just that they are designed for a specific purpose.
We'll search out every place a sick, twisted, solitary misfit might run to! -- I'll start with Radio Shack.
 
The following users thanked this post: ricko_uk

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #9 on: June 09, 2020, 10:45:29 pm »
Hi,
I need to protect a small experimental space so that it is immune to external noise as much as possible. I had few ideas which then raised a few questions:

1) what is the best material? Copper mesh, metal mesh, solid copper, solid steel, other?

2) is placing a Faraday cage inside another more effective? Or would it create some capacitor (or other) undesired effect?

3) if yes to (2) above should they be connected to each other? Of would connecting each other create some waves bouncing around inside between them and create other effects that ultimately affect the inner most space?

4) should I be grounding it or leave it floating? Or perhaps if (2) is a better solution should I ground perhaps the external one and leave floating the inner one?

Thank you :)

This is really an EMI/EMC issue. Many small details contribute to the overall efficacy of such enclosures.

It would probably help to read one or more of the standard textbooks on the topic. One that has been through several revisions over the decades is Ralph Morrison's "Grounding and Shielding Techniques in Instrumentation". That might help, and if not there are others.

It is at times like this that the demise of decent technical bookshops is felt. Once upon a time it would have been possible to go and skim a couple of books, then buy the one that looked most suitable.
go to a university Library
 
The following users thanked this post: ricko_uk

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #10 on: June 09, 2020, 10:51:06 pm »
Most microwave ovens are not great.  I've tested a cell phone inside a microwave and it still works.  They are made to be cheap and good enough for purpose not a lab grade screening cage.

Anyway for low frequency your limit will be magnetic shielding.  You will need a magnetic material like iron, steel, or mumetal to get good attenuation for Hz range magnet fields and it will be very hard to get good low frequency magnetic screening over a large volume.  Low frequency electric fields are easy to screen.  For high frequency the limiting factor will be your signal feedthroughs and the seams -- especially the door.  Real screenrooms usually have berillium copper  finger springs all the way around the door and compress them with a lot of force.  EMI gasket material is an easier alternative that can also be effective but may wear out with use and stop making good contact.  Springs are good because they scrape away oxide and contamination each time they are used to make sure the contact is good.
Good point.  i have used steel enclosures for EMP systems.  But you need to treat the doors.  For instance flame sprayed zinc.  The steel needs to have continuous welds at joints. 
 
The following users thanked this post: ricko_uk

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Faraday cage, shielding and grounding
« Reply #11 on: June 09, 2020, 11:23:24 pm »

Thank you all,

1) assuming a cylindrical shaped Faraday cage, and instead of a door it has a lid like one of those round metal chocolates boxes. Does that lid that "locks" into the main part of the box by "interference fit" provides tight enough of a seal or not?

2) does the metal thickness matter? If so, how does it affect the shielding?

3) I assume using one of those metal chocolate boxes is not good enough. What are the main reasons?

Thank you
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21231
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Faraday cage, shielding and grounding
« Reply #12 on: June 09, 2020, 11:46:35 pm »
Hi,
I need to protect a small experimental space so that it is immune to external noise as much as possible. I had few ideas which then raised a few questions:

1) what is the best material? Copper mesh, metal mesh, solid copper, solid steel, other?

2) is placing a Faraday cage inside another more effective? Or would it create some capacitor (or other) undesired effect?

3) if yes to (2) above should they be connected to each other? Of would connecting each other create some waves bouncing around inside between them and create other effects that ultimately affect the inner most space?

4) should I be grounding it or leave it floating? Or perhaps if (2) is a better solution should I ground perhaps the external one and leave floating the inner one?

Thank you :)

This is really an EMI/EMC issue. Many small details contribute to the overall efficacy of such enclosures.

It would probably help to read one or more of the standard textbooks on the topic. One that has been through several revisions over the decades is Ralph Morrison's "Grounding and Shielding Techniques in Instrumentation". That might help, and if not there are others.

It is at times like this that the demise of decent technical bookshops is felt. Once upon a time it would have been possible to go and skim a couple of books, then buy the one that looked most suitable.
go to a university Library

That is significantly more difficult than when I was a kid.

But then when I was a kid I used Downing Street in London as a short cut, walking right past No10. Good luck with that now :(
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: ricko_uk

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #13 on: June 10, 2020, 03:52:28 am »

Thank you all,

1) assuming a cylindrical shaped Faraday cage, and instead of a door it has a lid like one of those round metal chocolates boxes. Does that lid that "locks" into the main part of the box by "interference fit" provides tight enough of a seal or not?

2) does the metal thickness matter? If so, how does it affect the shielding?

3) I assume using one of those metal chocolate boxes is not good enough. What are the main reasons?

Thank you
There needs to be pressure on the contact.  A flat flange like used on a water pipe would be better.  Like a top hat.  I assume this is for an experiment and not for a product.  You could have a slip fit.  Use finger stock EMI gasket. 
 
The following users thanked this post: ricko_uk

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9323
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Faraday cage, shielding and grounding
« Reply #14 on: June 10, 2020, 04:55:23 am »
Most microwave ovens are not great.  I've tested a cell phone inside a microwave and it still works.  They are made to be cheap and good enough for purpose not a lab grade screening cage.

Microwave ovens are designed to contain the microwaves, not all random RF fields.  The door seals are usually a resonant gap that provides a virtual short-circuit at the specific microwave frequency used (2.45 GHz).  While this gap works very well at this frequency, it will not provide much if any attenuation at other frequencies.

So it's not a matter of cheap, it's just that they are designed for a specific purpose.
Some don't even shield enough to completely block Wifi.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 
The following users thanked this post: ricko_uk

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21231
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Faraday cage, shielding and grounding
« Reply #15 on: June 10, 2020, 07:10:09 am »
Most microwave ovens are not great.  I've tested a cell phone inside a microwave and it still works.  They are made to be cheap and good enough for purpose not a lab grade screening cage.

Microwave ovens are designed to contain the microwaves, not all random RF fields.  The door seals are usually a resonant gap that provides a virtual short-circuit at the specific microwave frequency used (2.45 GHz).  While this gap works very well at this frequency, it will not provide much if any attenuation at other frequencies.

So it's not a matter of cheap, it's just that they are designed for a specific purpose.
Some don't even shield enough to completely block Wifi.


After a quick glance at the production quality of that video, I assumed that also reflected on the technical content. So I decided I couldn't be bothered to spend 7 minutes of my life looking at it.

Last time I looked, in the early 90s (just as WLANs were becoming practical and before 802.11), domestic microwave ovens could leak up to 1W. Obviously that is enough to affect some WiFi comms, although the effect will be highly dependent on the geometry.

It it certainly enough that I don't get my corneas close to operating microwave ovens that I don't know have been well treated throughout their lives.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: ricko_uk

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Faraday cage, shielding and grounding
« Reply #16 on: June 10, 2020, 01:22:16 pm »
Thank you all for all the infos :)
 

Offline pwlps

  • Frequent Contributor
  • **
  • Posts: 372
  • Country: fr
Re: Faraday cage, shielding and grounding
« Reply #17 on: June 11, 2020, 11:14:02 am »

Thank you all,

1) assuming a cylindrical shaped Faraday cage, and instead of a door it has a lid like one of those round metal chocolates boxes. Does that lid that "locks" into the main part of the box by "interference fit" provides tight enough of a seal or not?

2) does the metal thickness matter? If so, how does it affect the shielding?

3) I assume using one of those metal chocolate boxes is not good enough. What are the main reasons?

Thank you

The thickness does matter : it has to be reasonably greater than the skin depth (https://en.wikipedia.org/wiki/Skin_effect). Otherwise you still have electrostatic shielding protecting against capacitive coupling (Faraday box) but you are not immune to noise coming via inductive (magnetic) coupling. Since the skin depth varies like 1/sqrt(f) magnetic shielding at low frequencies becomes difficult. This is well known in audio industry, simple coaxial cables (like RCA cables) are not sufficient at low frequencies this is why professional-grade equipment use differential transmission (XLR cables). As far as I know for very low frequencies a magnetic shielding using a mu-metal may be used (but I'm not specialist in this domain, I don't know the details).
« Last Edit: June 11, 2020, 11:23:06 am by pwlps »
 

Offline graybeard

  • Frequent Contributor
  • **
  • Posts: 443
  • Country: us
  • Consulting III-V RF/mixed signal/device engineer
    • Chris Grossman
Re: Faraday cage, shielding and grounding
« Reply #18 on: June 12, 2020, 10:26:19 pm »
I have used industrial emc enclosures from Rose Enclosures.  They are made of a proprietary copper-aluminum alloy that seems to avoid oxidation at the mating surface, yet is easy to drill or machine.

I have never attempted to measure their attenuation vs. frequency, but they have always been quiet enough for me to make the measurements I need.  I have used them to measure currents in Si and III-V devices down to the 10s of fAs.

Here is one I am working on now to use with my HP4145B.  I'm waiting on some stubby drills to finish it because it does not fit under my drill press with full sized drills.



Here is the inside of one I made 35 years ago.  I don't remember what the brand of the RF absorbent material is I used to coat the inside was, but it was put there to reduce resonances that were corrupting my measurements.



Here you can see that the gasket they provide is conductive.  That conductive gasket plus the lip on the lid helps with shielding.



. . . Chris
« Last Edit: June 12, 2020, 10:31:42 pm by graybeard »
 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #19 on: June 12, 2020, 11:11:58 pm »
I have used industrial emc enclosures from Rose Enclosures.  They are made of a proprietary copper-aluminum alloy that seems to avoid oxidation at the mating surface, yet is easy to drill or machine.

I have never attempted to measure their attenuation vs. frequency, but they have always been quiet enough for me to make the measurements I need.  I have used them to measure currents in Si and III-V devices down to the 10s of fAs.

Here is one I am working on now to use with my HP4145B.  I'm waiting on some stubby drills to finish it because it does not fit under my drill press with full sized drills.



Here is the inside of one I made 35 years ago.  I don't remember what the brand of the RF absorbent material is I used to coat the inside was, but it was put there to reduce resonances that were corrupting my measurements.



Here you can see that the gasket they provide is conductive.  That conductive gasket plus the lip on the lid helps with shielding.



. . . Chris
How does that compare to using chem filmed aluminum?
 

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Faraday cage, shielding and grounding
« Reply #20 on: June 12, 2020, 11:30:14 pm »
Thank you Greybeard! Very interesting, the conductive gasket too, that could be a possible solution!

Ahbushnell,
what is chem filmed aluminium? I googled it and it looks like some process, is it possible to somehow apply it to a standard enclosure too?

Thank you :)
 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 751
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #21 on: June 13, 2020, 07:04:46 pm »
Thank you Greybeard! Very interesting, the conductive gasket too, that could be a possible solution!

Ahbushnell,
what is chem filmed aluminium? I googled it and it looks like some process, is it possible to somehow apply it to a standard enclosure too?

Thank you :)
chem film is slang.
Trade name is Alodine.  You can get clear and yellow.  I like yellow because it shows it's been treated.  Plus it looks better.  :) 

You can buy kits and do it yourself or have it done.  We normally spec the coating the sheet metal house. 

https://usglassmag.com/bonding/alodine-aluminum-the-brand-versus-the-finish/


 
The following users thanked this post: ricko_uk

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Faraday cage, shielding and grounding
« Reply #22 on: June 13, 2020, 09:44:35 pm »
Thank you ahbushnell :)
 

Offline rhb

  • Super Contributor
  • ***
  • Posts: 3518
  • Country: us
Re: Faraday cage, shielding and grounding
« Reply #23 on: June 13, 2020, 11:49:35 pm »
There are two issues:  electric field and magnetic field.  Copper will not block magnetic fields whereas low carbon steel sheet is fairly effective above 1 MHz.  Screen drops in effectiveness as the frequency goes up. Low frequency magnetic fields require mumetal which is quite expensive as it is nearly pure nickel.  Galvanized steel sheet is a good enough conductor to be effective for the electric field.

A friend who is a retired  level II certified TEMPEST tech told me that the commercial rooms he dealt with used plywood with galvanized steel sheet on both sides and a custom rib that held the panels together with bolts every few inches.

I'm building a lab room approximately 15' x 11' which will have all the surfaces covered with 26 gauge galvanized steel sheet riveted to metal studs with all seams soldered.  The door will use metal weather stripping on  all edges.

For a small shielded enclosure I spot welded piano hinge on a pair of steel radio chassis along with a trunk latch.  Later it will get a layer of thermal insulation and a Peltier junction to control the temperature.

https://www.eevblog.com/forum/metrology/shielded-test-chamber/msg1471450/#msg1471450

A lot depends upon how big you need the chamber to be, whether there is significant heat being generated that needs to be removed and what type and  level of environmental noise you have.

I lived a mile or two from a 250 kW FM transmitter when I was in grad school.  Drove me nuts!  While I don't have that problem where I live now, it made me want a shielded room.

A reasonably inexpensive option for a quiet bench would be a 5 sided steel box holding your test gear and a copper screened front panel that allowed you to see the displays and provide air circulation.

You should make or buy an H field probe.  Dave did a video on them and there are some really cheap ones from China.  There is *no* justification for the $$$$ versions.  It's nothing special, just a shielded loop so it only responds to magnetic fields.  Combine it with an SDR to do spectrum analysis.

For a small low cost enclosure with good magnetic shielding consider a large diameter section of threaded steel pipe with malleable iron caps.  Thoroughly annealed if possible.

You will need to *very* carefully filter all lines passing through the walls.  The details of that depend very much on the signal. For fast signals make opto-isolators that send the signal through a hole in the enclosure and then put a metal cover over both sides.

It can get very gnarly.  So measure often and keep at it.

Have Fun!
Reg


 
The following users thanked this post: ricko_uk

Offline ricko_ukTopic starter

  • Super Contributor
  • ***
  • Posts: 1158
  • Country: gb
Re: Faraday cage, shielding and grounding
« Reply #24 on: June 17, 2020, 04:57:49 pm »
Thank you Reg and all :)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf