rf shielding problems.

[Rob]

Hi to all.
I am trying to make a shielded enclosure to test 403 and 433Mhz
transmitters and receivers.Low power stuff(few mW) .Garage door openers
etc.
I have put a xmtr in a die case aluminuim box( 12 * 6 * 3 cm) , With no
openings. I got
pretty good attenuation of the signal. I am using the Micrel MICrf011
receiver chips , and using the AGC as a basic signal strength meter.
Seems to work ok. With the box closed tight I could only just detect the
signal.
I now needed to bring some wires into the box to supply sepperate power ,
turn on buttons etc. I have brought the wires into the box via a DB9
connector (male) mounted in the side of the box. The box has been machined
to only just fit the connector.(tight fit).
With this arrangement quite a lot of RF got out.
The wires are obviously acting as antennas and bringing the signal out of
the box.
I then put(soldered) 100pF surface mount caps from each used pin to the
chassis of the connector , which is in contact with the metal box.I did
this on both the male(machined into the box) and the female ( used to
attach wires).
The signal was attenuated quite a lot , but not enough for my liking.
I tried earthing the box , no difference. I get the feeling that would not
make a difference anyway , as a long wire (1M) to earth is a fat inductor
at 433Mhz. Tried taking the negative of the xmitter cct in the box to the
chassis , made thing worse I think :0(
Is there much more I can do , or am I barking up the wrong tree entirely!!

Any help would be appreciated.
Cheers
Rob

 

[Don Pearce]

Hi to all.
I am trying to make a shielded enclosure to test 403 and 433Mhz
transmitters and receivers.Low power stuff(few mW) .Garage door openers
etc.
I have put a xmtr in a die case aluminuim box( 12 * 6 * 3 cm) , With no
openings. I got
pretty good attenuation of the signal. I am using the Micrel MICrf011
receiver chips , and using the AGC as a basic signal strength meter.
Seems to work ok. With the box closed tight I could only just detect the
signal.
I now needed to bring some wires into the box to supply sepperate power ,
turn on buttons etc. I have brought the wires into the box via a DB9
connector (male) mounted in the side of the box. The box has been machined
to only just fit the connector.(tight fit).
With this arrangement quite a lot of RF got out.
The wires are obviously acting as antennas and bringing the signal out of
the box.
I then put(soldered) 100pF surface mount caps from each used pin to the
chassis of the connector , which is in contact with the metal box.I did
this on both the male(machined into the box) and the female ( used to
attach wires).
The signal was attenuated quite a lot , but not enough for my liking.
I tried earthing the box , no difference. I get the feeling that would not
make a difference anyway , as a long wire (1M) to earth is a fat inductor
at 433Mhz. Tried taking the negative of the xmitter cct in the box to the
chassis , made thing worse I think :0(
Is there much more I can do , or am I barking up the wrong tree entirely!!

Any help would be appreciated.
Cheers
Rob

Replace the connector with proper feed-thru capacitors and put ferrite
beads on the leads inside and outside the box. That should do it. If
it doesn't, put an extra wall inside the box with another feedthru,
and a decent wound inductor in the power line between the feedthrus.

d

Pearce Consulting
http://www.pearce.uk.com

 

[Dbowey]

Rob posted:
<< I am trying to make a shielded enclosure to test 403 and 433Mhz
transmitters and receivers.Low power stuff(few mW) .Garage door openers
etc.
I have put a xmtr in a die case aluminuim box( 12 * 6 * 3 cm) , With no
openings. I got
pretty good attenuation of the signal. I am using the Micrel MICrf011
receiver chips , and using the AGC as a basic signal strength meter.
Seems to work ok. With the box closed tight I could only just detect the
signal.
I now needed to bring some wires into the box to supply sepperate power ,
turn on buttons etc. I have brought the wires into the box via a DB9
connector (male) mounted in the side of the box. The box has been machined
to only just fit the connector.(tight fit).
With this arrangement quite a lot of RF got out.
The wires are obviously acting as antennas and bringing the signal out of
the box.
I then put(soldered) 100pF surface mount caps from each used pin to the
chassis of the connector , which is in contact with the metal box.I did
this on both the male(machined into the box) and the female ( used to
attach wires).
The signal was attenuated quite a lot , but not enough for my liking.
I tried earthing the box , no difference. I get the feeling that would not
make a difference anyway , as a long wire (1M) to earth is a fat inductor
at 433Mhz. Tried taking the negative of the xmitter cct in the box to the
chassis , made thing worse I think :0(
Is there much more I can do , or am I barking up the wrong tree entirely!!
I would do two things - Use feed-through capacitors on the box/leads if there
is room, and use a ferrite bead or two on each lead as close to the outside of
the box as you can place them.

Don

 

[Norm Dresner]

Hi to all.
I am trying to make a shielded enclosure to test 403 and 433Mhz
transmitters and receivers.Low power stuff(few mW) .Garage door openers
etc.
I have put a xmtr in a die case aluminuim box( 12 * 6 * 3 cm) , With no
openings. I got
pretty good attenuation of the signal. I am using the Micrel MICrf011
receiver chips , and using the AGC as a basic signal strength meter.
Seems to work ok. With the box closed tight I could only just detect the
signal.
I now needed to bring some wires into the box to supply sepperate power ,
turn on buttons etc. I have brought the wires into the box via a DB9
connector (male) mounted in the side of the box. The box has been machined
to only just fit the connector.(tight fit).
With this arrangement quite a lot of RF got out.
The wires are obviously acting as antennas and bringing the signal out of
the box.
I then put(soldered) 100pF surface mount caps from each used pin to the
chassis of the connector , which is in contact with the metal box.I did
this on both the male(machined into the box) and the female ( used to
attach wires).
The signal was attenuated quite a lot , but not enough for my liking.
I tried earthing the box , no difference. I get the feeling that would not
make a difference anyway , as a long wire (1M) to earth is a fat inductor
at 433Mhz. Tried taking the negative of the xmitter cct in the box to the
chassis , made thing worse I think :0(
Is there much more I can do , or am I barking up the wrong tree entirely!!

Any help would be appreciated.
Cheers
Rob

There are D-Sub-mini connectors with filter pins that do exactly what you
want. Not cheap; a couple of feedthrus and ferrite beads would be more cost
effective, though significantly more assembly time. If it's a one-off, I'd
go the feedthru route.

Norm

 

[Joerg]

Hi Norm,

There are D-Sub-mini connectors with filter pins that do exactly what you
want. Not cheap; a couple of feedthrus and ferrite beads would be more cost
effective, though significantly more assembly time. If it's a one-off, I'd
go the feedthru route.

These actually have a ferrite plate with as many holes as there are
pins. Still he'd need chip caps in addition to that for some
differential mode suppression versus chassis.

Regards, Joerg

 

[Joerg]

Hi Rob,

You got many hints already. Just one more. A DB9 isn't the optimal choice here. Try coax connectors, even if it was just am RCA or something that would still be better than DB9. Now run a chip cap right from center to ground and a ferrite bead before that.

If it comes down to the wire there is another option: The good old feedthrough caps. You have to use a tin can for that if you want to use the cheaper soldered versions. Ham radio shops should have these.


Regards, Joerg

 

[Rob]

Hi Norm,

These actually have a ferrite plate with as many holes as there are
pins. Still he'd need chip caps in addition to that for some
differential mode suppression versus chassis.

Regards, Joerg

Thanks for the tips guys. Getting some feed-through caps and ferrite beads
next week.Hold thumbs!!
Cheers
Rob

 

[Joerg]

Hi Rob,

Once it works place a dab of soft but durable glue on the beads. Otherwise they produce a nasty rattle and sometimes even shatter.

Good luck, Joerg