EMRFD Message Archive 7882

Message Date From Subject
7882 2012-10-20 20:12:29 Ray Shielded enclosure effectiveness
I've used the Hammond cast aluminum boxes in the past to shield various circuits. I've also used double sided copper clad to build boxes to shield circuits. Has anyone performed any tests to determine if the aluminum boxes are better at shielding than the copper clad ones? Would one form be better than the other at different frequency ranges?
7883 2012-10-21 04:51:36 Alan Melia Shielded enclosure effectiveness
Hi Ray, I think Ian White GM3SEK has something on his web site.

As far as I am aware it is not the metal-material that is too much of a
problem it is more the "holes", access leads, and grounding that determine
the effectiveness. It is interesting that the UK microwave gang use 0.5mm
tinplate extensively. Its main point beyond relative cheapness is the
ability to solder all round pcb ground planes, and solder coax connector
flanges. At microwaves this has more to do with improved stability than
leakage.

Size of holes and length of "slots" between lid fixing screws are the most
important parameters. Look at the multiple cover bolts on good sig-gen
enclosures and the bolts are about 1/2inch apart!! coax is wound through
common-mode chokes etc etc.

Alan
G3NYK
7884 2012-10-21 06:04:36 Ashhar Farhan Re: Shielded enclosure effectiveness
Ray,
Using double sided pcb material is a little tricky. Both side should
be held at ground and with multiple passhtroughs.
Otherwise, ir has quite the opposite behaviour. I noticed this when I
had to isolate a vhf local oscillator for a sweep generator.
The best materil turned out to be thin copper. You can solder it from
any side and then solder the lid inside as well. For isolation of dc
lines or control lines, a smaller box inside the box can be used. It
was suggested by someone on emrfd. Bring the dc and controls into the
smaller box from outisde, bypass them on ingress as well as egress
(the wires pass through to the main comparment).
These boxes can't match the mechanical stability of a die cast box.
But the rf isolation can be excellent!

On 10/21/12, Ray <rcbuckiii@yahoo.com> wrote:
> I've used the Hammond cast aluminum boxes in the past to shield various
> circuits. I've also used double sided copper clad to build boxes to shield
> circuits. Has anyone performed any tests to determine if the aluminum boxes
> are better at shielding than the copper clad ones? Would one form be better
> than the other at different frequency ranges?
>
>

--
Sent from my mobile device
7885 2012-10-21 06:09:18 Ashhar Farhan Re: Shielded enclosure effectiveness
I forgot, there was a paper somewhere that gave details about the box
sizes viz the frequency of operation. This applies only to UHF and
microwaves, I guess.
S53MV also mentions about microwave absorbing foam. There is some info
on his site about rf shielding in the context of his 1750MHz spec an.
- Farhan

On 10/21/12, Ashhar Farhan <farhanbox@gmail.com> wrote:
> Ray,
> Using double sided pcb material is a little tricky. Both side should
> be held at ground and with multiple passhtroughs.
> Otherwise, ir has quite the opposite behaviour. I noticed this when I
> had to isolate a vhf local oscillator for a sweep generator.
> The best materil turned out to be thin copper. You can solder it from
> any side and then solder the lid inside as well. For isolation of dc
> lines or control lines, a smaller box inside the box can be used. It
> was suggested by someone on emrfd. Bring the dc and controls into the
> smaller box from outisde, bypass them on ingress as well as egress
> (the wires pass through to the main comparment).
> These boxes can't match the mechanical stability of a die cast box.
> But the rf isolation can be excellent!
>
> On 10/21/12, Ray <rcbuckiii@yahoo.com> wrote:
>> I've used the Hammond cast aluminum boxes in the past to shield various
>> circuits. I've also used double sided copper clad to build boxes to
>> shield
>> circuits. Has anyone performed any tests to determine if the aluminum
>> boxes
>> are better at shielding than the copper clad ones? Would one form be
>> better
>> than the other at different frequency ranges?
>>
>>
>
> --
> Sent from my mobile device
>

--
Sent from my mobile device
7886 2012-10-21 06:34:44 Hector Pascal Re: Shielded enclosure effectiveness
With DS PCB it isn't difficult to make a continuous/contiguous solder metal
surface internally on sides 1-5 to form an open box. It's solder seaming
the internal surface of the 6th side that's the problem!

Using thin copper foil, I wrap and solder the exposed inside edges of the
box opening, to the outside copper edge. I then cut side 6 slightly too
large in each dimension, and solder seam the slightly exposed internal
surface of this lid to the copper foil wrappings.

The isolation obtained is usually good enough for all but the most critical
applications, but is often spoiled by fitting coaxial connectors, unless
their holes are similarly wrapped and soldered to the box

However, having said all this, the convenience of the diecast box lid is
unbeatable. Has anybody solved this removable lid problem in PCB box
construction, and still maintained good isolation?

David VK6JT


[Non-text portions of this message have been removed]
7888 2012-10-21 09:15:09 Alex P Re: Shielded enclosure effectiveness
Standard EMI/EMC design procedure is to use a conductive elastomeric gasket. They can be purchased with copper, nickel, silver filling, etc., in a silicone rubber or various other elastomeric matrix. You can also use conductive Be/Cu fingerstock, or spiral gasket as another alternative. There are other remediation methods available besides the foregoing. Electric Field Shielding Effectiveness (SE) depends on a variety of parameters, but you can get these materials/items that typically give you SE in the range of 60+ dB as high as 2+ GHz but it depends
7890 2012-10-21 11:58:05 Ray Re: Shielded enclosure effectiveness
Thanks for all the replies. It appears the DS board material is ample for what I do. I do not need to go any higher than 100 MHz as I don't do any VHF or UHF stuff. I've never operated 2 meters and only operated 6 meters for a year or so back in the mid-60s.

What I have been doing is similar to what David described except I don't use copper foil to connect the inside surfaces to the outside sufaces. I solder seam the inside surfaces on the five pieces with a complete solder bead. I depend on the connectors and feed-through caps to connect the inside surfaces to the outside surfaces. On the sides that don't have connectors I use four 6-32 screws, lock washers, and nuts near the corners to connect inside and outside surfaces. The copper foil would probably be more effective to connect the inside and outside surfaces.

The sixth side is about 1/4 inch larger than the box and soldered completely around the box on the four sides. This side has 4 holes that are used to mount the box to the main chassis. The mounting screws connect the top surface to the bottom surface of this side.

For my W7ZOI spectrum analyzer I used the diecast boxes as described in the article. I didn't feel that PCB material would have provided effective shielding for the analyzer.

Ray
AB7HE


7891 2012-10-21 13:43:57 Kerry Re: Shielded enclosure effectiveness
I had an interesting experience recently.

I wanted a very high gain amplifier so I bought a nice little two-MMIC kit from Mini Kits here in VK.

I have powerful FM radio & TV transmitters within a few miles so shielding was obviously required.

I put the PCB in a Hammond case;

http://i48.tinypic.com/jtw3gn.jpg

and tested it.

Here the case lid is off;

http://i47.tinypic.com/2z8xis9.jpg

The test signal is at 500 MHz; there is about 40 dB of gain, just what I wanted. The FM radio stations can be seen down in their ca.100 MHz band and the four TV stations can be seen up higher.

Then I fitted the case lid;

http://i49.tinypic.com/afalfp.jpg

Excellent shielding; the radio & TV stations have gone.

But so has the gain; it's down to about 10 dB!

It's never a good idea to use a case larger than necessary; there were clearly some resonances going-on inside the Hammond case.

I solved the problem by using a PCB case fitted around the amplifier PCB;

http://i48.tinypic.com/iqhqgw.jpg

(The lid is off the PCB box in the above photo).

With this "double-shielding", gain returns and the nasties are kept out; I don't have a sweep of the final gain but it's shown on the label I made;

http://i46.tinypic.com/s0x88y.jpg

I prefer die-cast cases for their appearance when painted but I occasionally use a PCB case as in this amplifier;

http://i48.tinypic.com/70xhnd.jpg

http://i47.tinypic.com/350rxc6.jpg

Single-sided PCB works fine, especially if seam-soldered; even if, as in the above example, the top & bottom are just lids screwed-on, shielding is very good.

Wes has discussed the effectiveness of PCB boxes in EMRFD and elsewhere as I recall.

One other thing, not directly connected with boxes but it might be interesting; here is a good way to make "ugly" circuits inside a box;

http://i45.tinypic.com/1198pkn.jpg

Soldering down inside a box can be difficult and this method gives a lot more "elbow-room" when building the circuit.

The PCB "chassis" is held in the box by the two BNCs screwed into nuts soldered to the end plates. The circuit can be built & tested on the chassis and, when completed, the two BNCs are unsoldered & removed, the chassis dropped into the box and the BNCs screwed in and re-soldered.

This, incidentally, is a two-stage Norton-feedback amplifier; there is a PCB shield between the two stages.

When finished, no-one would know that it's "ugly" inside;

http://i48.tinypic.com/vql35h.jpg

Kerry VK2TIL.
7896 2012-10-21 15:50:14 Jim Miller Re: Shielded enclosure effectiveness
I like the die cast boxes. The lid has a tongue that fits in a groove in
the box to make a waveguide below cutoff around the periphery.

No need for a gasket.

jim ab3cv


[Non-text portions of this message have been removed]
7901 2012-10-21 19:51:26 rlramirez77 Re: Shielded enclosure effectiveness
Have a good source for the smaller ones Jim? In the range of 1.5"w x 2"l x 1"d for example.

Russ
K0WFS

7902 2012-10-21 19:58:42 Jim Miller Re: Shielded enclosure effectiveness
my memory was faulty. it's an L shaped fit rather than tongue in groove.
however there's likely little difference. the hammond ones are available
from places like mouser and digikey. lots of them usually in stock.
http://www.hammondmfg.com/dwg.htm

i bought a bunch of them to build my W7ZOI spectrum analyzer...stilll
unfiinished in messy basement...

jim ab3cv


[Non-text portions of this message have been removed]