i've been building all my latest stuff with wood enclosures and wonder if i'm making them vulnerable to rfi. i've looked at copper lining but at that price, i may as well get a metal chassis and decorate it with hardwood.
is there anything more economical or do i even need to bother with wooden enclosure?
thanks
is there anything more economical or do i even need to bother with wooden enclosure?
thanks
I use many of my amplifiers "on test" for extended periods running into weeks and even months without any chassis.
I don't get interference that I can hear.
Because I try to minimise the loop areas of all the circuits
I don't get interference that I can hear.
Because I try to minimise the loop areas of all the circuits
Copper foil lining should not be too expensive. But the best solution is
steel or aluminum chassis. Wood enclosure will serve only decoration purpose.
steel or aluminum chassis. Wood enclosure will serve only decoration purpose.
I'm against wood enclosures due to the RF issues and fire hazard. UL and CE certainly wouldn't let you get away with it. That said, lots of old test equipment had wood enclosures, like the classics from General Radio Corp. They typically lined them with copper. I think you'd have just as good a result using aluminum roofing flashing. It's cheap and cuts with scissors or a paper cutter. The only downside is you can't solder to it. Actually, you can with the right flux and technique, but it's a PITA. Screws are easier.
Hammond makes many types, even with wood sides these days. Download the product guide and take a look.
You can buy used equipment and use those chassis. If certified products, try to follow how they did the wiring.
For nice but expensive front/back panels, Front Panel Express. There SW is nice to use just to plan things out/print/check.
You can buy used equipment and use those chassis. If certified products, try to follow how they did the wiring.
For nice but expensive front/back panels, Front Panel Express. There SW is nice to use just to plan things out/print/check.
I sometimes use budget pie or cake pans.
But being made of aluminum, making ground connections is more difficult.
But being made of aluminum, making ground connections is more difficult.
No one. To make a shield, you first need a good current conductor, then it will create a equipotential zone near it.
What it means, is that a variable magnetic field near a metal surface creates a current in it, then current re-generates a magnetic field such as to opposite to the original magnetic field. If thickness of metal is sufficiently big, the magnetic field cannot run inside the material, then it is said to be a shield. If too fine, magnetic field continues in the other face, and then a no shielding effect appears.
But, which is the boundary to be "sufficient thickness"?. Well, there is a measure of the penetration inside the material, which is estimated in such a manner that the magnetic field decreases to 1/e times (about 38%) in the non incident surface. And as can you see, it will depend on the conductivity of the material and the frequency of the incident magnetic variable field.
Some similar effect appears with electric fields, acting the armature of a capacitor grounded, and finishing in it the lines of force of it.
Penetration in copper was about 85/sqrt(f) with f in hertz and penetration in mm (If I remember well).
For more accurate data, search the "proxi effect" in transformers and inductors or http://en.wikipedia.org/wiki/Skin_effect. Your shielding material will need to have a thickness greater than the penetration for good shielding effect.
Sorry if my English is not too good, perhaps DF96 can be more accurate at this respect.
What it means, is that a variable magnetic field near a metal surface creates a current in it, then current re-generates a magnetic field such as to opposite to the original magnetic field. If thickness of metal is sufficiently big, the magnetic field cannot run inside the material, then it is said to be a shield. If too fine, magnetic field continues in the other face, and then a no shielding effect appears.
But, which is the boundary to be "sufficient thickness"?. Well, there is a measure of the penetration inside the material, which is estimated in such a manner that the magnetic field decreases to 1/e times (about 38%) in the non incident surface. And as can you see, it will depend on the conductivity of the material and the frequency of the incident magnetic variable field.
Some similar effect appears with electric fields, acting the armature of a capacitor grounded, and finishing in it the lines of force of it.
Penetration in copper was about 85/sqrt(f) with f in hertz and penetration in mm (If I remember well).
For more accurate data, search the "proxi effect" in transformers and inductors or http://en.wikipedia.org/wiki/Skin_effect. Your shielding material will need to have a thickness greater than the penetration for good shielding effect.
Sorry if my English is not too good, perhaps DF96 can be more accurate at this respect.
Last edited:
For magnetic field rejection you need a magnetic material, mu metal the best, and small loops, for electric field rejection you need a grounded conductive shield. Steel works for both, but its not the best for either. Magnetic noise is usually not a problem in residential unless your near a transformer or motor.
You need to be clear what you are shielding against.
For low frequency magnetic fields, including static fields, you need mu metal - but this is difficult to work with as it needs re-annealing after any bending or drilling etc.
For high frequency magnetic fields you need a complete metal shield with no seams or holes, except some which are much smaller than the wavelength involved. Thickness should be at least several skin depths.
For low frequency electric fields you need a metal screen, but it can have holes or small gaps. For high frequency electric fields the requirement is the same as for high frequency magnetic fields.
For audio and 50/60Hz hum what this basically means is that you need small circuit loops to reduce magnetic coupling because you can't shield from it. Electric fields can be reduced by quite crude shields with lots of gaps.
For RF you need a continuous metal shield with no gaps or seams. You also need to be careful about power and signal entry/exit points as these can break the shield and carry RF in/out. Very difficult to achieve this without a metal case.
Wood of any realistic thickness is useless for this. Looks nice, but electromagnetically it almost isn't there at all.
For low frequency magnetic fields, including static fields, you need mu metal - but this is difficult to work with as it needs re-annealing after any bending or drilling etc.
For high frequency magnetic fields you need a complete metal shield with no seams or holes, except some which are much smaller than the wavelength involved. Thickness should be at least several skin depths.
For low frequency electric fields you need a metal screen, but it can have holes or small gaps. For high frequency electric fields the requirement is the same as for high frequency magnetic fields.
For audio and 50/60Hz hum what this basically means is that you need small circuit loops to reduce magnetic coupling because you can't shield from it. Electric fields can be reduced by quite crude shields with lots of gaps.
For RF you need a continuous metal shield with no gaps or seams. You also need to be careful about power and signal entry/exit points as these can break the shield and carry RF in/out. Very difficult to achieve this without a metal case.
Wood of any realistic thickness is useless for this. Looks nice, but electromagnetically it almost isn't there at all.
- Status
- Not open for further replies.