I have been digging around in the archives looking for some information on magnetic shielding with non ferrous materials. I saw varying opinions about its effectiveness. Looking for info on the internets in general I came across this paper
http://www.chomerics.com/products/documents/emicat/pg192theory_of_emi.pdf
Reading through it seems that aluminum can be an effective magnetic shield if it is thick enough. Playing around with the nomographs at the end it looks to me like .25” of aluminum is about as absorptive as .009” of Mu Metal (about 6dB of absorption losses at 60 Hz) the reflection losses look to be about 20 dB but I gather these are to be ignored if you are only worried about the magnetic portion of the wave.
Does anyone have practical experience or a better grasp of the math that confirms what I think I see?
Thanks,
Marty
http://www.chomerics.com/products/documents/emicat/pg192theory_of_emi.pdf
Reading through it seems that aluminum can be an effective magnetic shield if it is thick enough. Playing around with the nomographs at the end it looks to me like .25” of aluminum is about as absorptive as .009” of Mu Metal (about 6dB of absorption losses at 60 Hz) the reflection losses look to be about 20 dB but I gather these are to be ignored if you are only worried about the magnetic portion of the wave.
Does anyone have practical experience or a better grasp of the math that confirms what I think I see?
Thanks,
Marty
not much (none)
looks like your on the right track ....
blowtorch photo -
An externally hosted image should be here but it was not working when we last tested it.
http://www.diyaudio.com/forums/solid-state/71189-john-curls-blowtorch-preamplifier.html#post810164
transformers mounted on one piece chassis -
An externally hosted image should be here but it was not working when we last tested it.
http://page11.auctions.yahoo.co.jp/jp/auction/n82058654
Last edited:
No grasp of the math, but I can confirm that thick aluminum does provide some shielding. I've used a piece between the power transformer and output transformer in tube projects. Noticeable reduction in mains hum. In my case, thickness was about 8mm. As good as 5mm mild steel.
Sheldon
you can turn transformers through 90 degrees to each other which helps stop the fields of the transformers effecting each other.
otherwise the metal work of the chassis is used mainly to block out RF and airborne hash from the small component connecting wires and pcb copper tracks.
there are few other components that are effected by low level magnetic fields such as (j type) wire wound resistors and some say valves and caps.
with the valves all you normally need to do is space them well and caps can be put below and spaced 1"+ from each other to stop them from over heating each other.
otherwise the metal work of the chassis is used mainly to block out RF and airborne hash from the small component connecting wires and pcb copper tracks.
there are few other components that are effected by low level magnetic fields such as (j type) wire wound resistors and some say valves and caps.
with the valves all you normally need to do is space them well and caps can be put below and spaced 1"+ from each other to stop them from over heating each other.
Well if I am following the paper I linked to it seems that it will just not as well as ferrous metals for a given thickness.
I have read others who say it will not work at all but I always had this nagging question. If highly conductive metals have no effect on a magnetic circuit why are they used as shorting rings in loudspeakers?
My personal experience is that it does work. The difference is clear with the plate between the transformers. It can be heard as the plate is moved into position, so it's not imaginary.
The material used was not pure aluminum. I believe it was T6.
Sheldon
Last edited:
Your personal experience would be correct. I did a google and found out that there is such a thing as lossy magnetic shielding.
Magnetic Shielding Solutions
"Lossy magnetic shielding depends on the eddy-current losses that occur within highly conductive materials (i.e., copper, aluminum, iron, steel, silicon-iron, etc.). When a conductive material is subjected to a time-varying (60 hertz) magnetic field, currents are induced within the material that flow in closed circular paths - perpendicular to the inducing field. According to Lenz's Law, these eddy-currents oppose the changes in the inducing field, so the magnetic fields produced by the circulating eddy- currents attempt to cancel the larger external inducing magnetic fields near the conductive surface, thereby generating a shielding effect."
I read somewhere else on the bottlehead forum that this was only about 5-6dB maximum (according to John Swenson)
http://www.bottlehead.com/smf/index.php/topic,70.msg290.html#msg290
Nothing close to mu-metals but it still works a little.
Magnetic Shielding Solutions
"Lossy magnetic shielding depends on the eddy-current losses that occur within highly conductive materials (i.e., copper, aluminum, iron, steel, silicon-iron, etc.). When a conductive material is subjected to a time-varying (60 hertz) magnetic field, currents are induced within the material that flow in closed circular paths - perpendicular to the inducing field. According to Lenz's Law, these eddy-currents oppose the changes in the inducing field, so the magnetic fields produced by the circulating eddy- currents attempt to cancel the larger external inducing magnetic fields near the conductive surface, thereby generating a shielding effect."
I read somewhere else on the bottlehead forum that this was only about 5-6dB maximum (according to John Swenson)
http://www.bottlehead.com/smf/index.php/topic,70.msg290.html#msg290
Nothing close to mu-metals but it still works a little.
I had to extend the thickness scale to get to ¼” so this may not be super accurate but a couple of things I take away from this
Aluminum works to absorb magnetic fields. 250 mils of aluminum is as effective as 6 mils of mu metal or about 100 mils of stainless all roughly equal 6dB of attenuation at 60 Hz. I didn’t see anything in the text that would lead me to believe multiple shields would have any unexpected effect with low impedance low frequency fields so stacking spaced plates should not be a problem.
Thanks for all of the replies, I am working on a layout for some sensitive tubes and output transformers and would rather keep the whole project on one chassis. I have plenty of aluminum and all of the equipment I need to fabricate so it seems like the easy way out for me. As it stands I have about 7” between the power transformer and the outputs with both being shielded ¼” aluminum cans. I have enough space for a couple of plates between the shields if required so hopefully that will work well enough.
Aluminum works to absorb magnetic fields. 250 mils of aluminum is as effective as 6 mils of mu metal or about 100 mils of stainless all roughly equal 6dB of attenuation at 60 Hz. I didn’t see anything in the text that would lead me to believe multiple shields would have any unexpected effect with low impedance low frequency fields so stacking spaced plates should not be a problem.
Thanks for all of the replies, I am working on a layout for some sensitive tubes and output transformers and would rather keep the whole project on one chassis. I have plenty of aluminum and all of the equipment I need to fabricate so it seems like the easy way out for me. As it stands I have about 7” between the power transformer and the outputs with both being shielded ¼” aluminum cans. I have enough space for a couple of plates between the shields if required so hopefully that will work well enough.
Attachments
Pure aluminum is not used often, as it is very soft. An example of aluminum alloy is 6063-T5, used in heat sinks. It contains the following elements:
Silicon,iron,copper,manganese,magnesium,chromium,zinc and titanium. All in small quantity's, however enough to change it's magnetic shielding capability.
Regards,
Eric
This absolutely does work and I've used aluminum plate for shielding very high gain amps in scientific equipment. In some cases it can be way more effective than mu-metal. The problem is you need a lot of thickness for it to be effective. 1/2 to 5/8 inch thick plate works well. 1/4 inch is pretty much useless. Your circuit needs everything else done correctly- grounding for example, but if there's some residual level of magnetic hum from transformer fields, this can eliminate it.
I ordered three sheets of this Ultraperm 80 a nickel-iron alloy (80.3% nickel, 14.3% iron, plus 14.3 %molybdenum). It should arrive next week. I have high hopes for it as I have a PP2A3 amp on a very small chassis and the induced noise from the PT, choke and rectifier are hard to eliminate.
SALE! Rare Ultraperm 80 Shielding Sheet-The Electronic Goldmine
SALE! Rare Ultraperm 80 Shielding Sheet-The Electronic Goldmine
use cheap Iron, star gnd, twisted pairs
air space and cheap Iron (low carbon steel) is likely better for most diy audio builders - magnetic materials "shield" by providing a low reluctance conductive path around the shielded area and so have to be sized to conduct the flux without saturating
with internal power supply its best to "short out" the source: you can reduce/intercept transformer leakage flux with a can/box of good magnetic conductor - see the old Lambda power supplies
then if there is still a problem with sensitive circuits use a 2nd box around the circuit
even "tin can" material can be an effective magnetic shield
simple distance is invaluable in magnetic shielding - the fields are dipole and fall off at a cubic rate - use external power supplies where practical
hi perm materials have special forming/annealing requirements and thin foils are easily saturated - they should be only used after cheap Iron has redirected the major interfering flux
I design strain gage amps at work and use welded steel boxes, external wall wart supply and twisted pair wiring - haven't needed exotic materials while using gains of 4000+ and looking at the results with 16 bit resolution
I'd guess except for circuits next to a line power transformer diy builder's main 50/60 Hz noise problems will be related to poor gnd design/layout
air space and cheap Iron (low carbon steel) is likely better for most diy audio builders - magnetic materials "shield" by providing a low reluctance conductive path around the shielded area and so have to be sized to conduct the flux without saturating
with internal power supply its best to "short out" the source: you can reduce/intercept transformer leakage flux with a can/box of good magnetic conductor - see the old Lambda power supplies
then if there is still a problem with sensitive circuits use a 2nd box around the circuit
even "tin can" material can be an effective magnetic shield
simple distance is invaluable in magnetic shielding - the fields are dipole and fall off at a cubic rate - use external power supplies where practical
hi perm materials have special forming/annealing requirements and thin foils are easily saturated - they should be only used after cheap Iron has redirected the major interfering flux
I design strain gage amps at work and use welded steel boxes, external wall wart supply and twisted pair wiring - haven't needed exotic materials while using gains of 4000+ and looking at the results with 16 bit resolution
I'd guess except for circuits next to a line power transformer diy builder's main 50/60 Hz noise problems will be related to poor gnd design/layout
Last edited:
Magnetic shielding
Hi guy's,
I have been looking around for a suitable material to magnetically shield T-amp air core inductors, I have read multiple posts today telling me of the benefits and the negatives of doing this.
From my search it appears that mu-metal is the ultimate product to use for this purpose, searching the net I have come up with a couple of possible sources:
1. ebay: PMT and Transformer MAGNETIC SHIELDING MAT'L- NEW - eBay Radiation Detectors Geigers, Safety Security, Industrial Supply MRO, Business Industrial. (end time 11-Dec-09 02:30:53 AEDST)
2. a web search gave me this: Engineering Kit
My question: The ebay item is a lot cheaper, the % of the alloys in the material appear to be very similar to that of Mu-metal, has anyone purchased any of this and tried it? I would welcome any comments on this and if you want, further positives and negatives of shielding chokes.
cheers.. Steve
Hi guy's,
I have been looking around for a suitable material to magnetically shield T-amp air core inductors, I have read multiple posts today telling me of the benefits and the negatives of doing this.
From my search it appears that mu-metal is the ultimate product to use for this purpose, searching the net I have come up with a couple of possible sources:
1. ebay: PMT and Transformer MAGNETIC SHIELDING MAT'L- NEW - eBay Radiation Detectors Geigers, Safety Security, Industrial Supply MRO, Business Industrial. (end time 11-Dec-09 02:30:53 AEDST)
2. a web search gave me this: Engineering Kit
My question: The ebay item is a lot cheaper, the % of the alloys in the material appear to be very similar to that of Mu-metal, has anyone purchased any of this and tried it? I would welcome any comments on this and if you want, further positives and negatives of shielding chokes.
cheers.. Steve
That eBay item comes from here:
SALE! Rare Ultraperm 80 Shielding Sheet-The Electronic Goldmine
Notice it is only $7.00 per sheet. I bought three sheets but haven't had a chance to use them yet.
SALE! Rare Ultraperm 80 Shielding Sheet-The Electronic Goldmine
Notice it is only $7.00 per sheet. I bought three sheets but haven't had a chance to use them yet.
What hasn't been mentioned in this thread is electromagnetic vs. magnetic shielding. DC magnetic shielding is what you want around your scope CRT so the electron beam flies straight. Electromagnetic shielding is what you want to keep nearby radio stations out of your amp.
At 60Hz and below, mild steel and mu-metal are orders of magnitude better than aluminum, as has been mentioned. If shielding is paramount, do the right thing and buy strips of permalloy, like those from Goldmine, and line your ordinary aluminum chassis with it. There is no need to use absurd thicknesses of aluminum except to demonstrate one's Freudian insufficiencies. 0.060" aluminum is plenty, easy to work, stiff enough to support transformers, and fairly priced.
At audio frequencies and up, aluminum and copper are better, because the field induces surface currents, reflecting rather than transmitting it. Aluminum is the metal of choice for my induction heater,
http://myweb.msoe.edu/williamstm/Images/Induction1103.jpg
because it is very effective at reflecting the 20-50kHz, high amperage fields present inside. If steel were used, it would be heated excessively by eddy currents and hysteresis loss. Aluminum runs cool.
Static magnetic fields have little effect on the average vacuum tube, after all we've been running our amps in the Earth's magnetic field for about a century with no ill effects. Use metal chassis to shield RFI, and use proper layout and wiring to exclude 60Hz.
Tim
At 60Hz and below, mild steel and mu-metal are orders of magnitude better than aluminum, as has been mentioned. If shielding is paramount, do the right thing and buy strips of permalloy, like those from Goldmine, and line your ordinary aluminum chassis with it. There is no need to use absurd thicknesses of aluminum except to demonstrate one's Freudian insufficiencies. 0.060" aluminum is plenty, easy to work, stiff enough to support transformers, and fairly priced.
At audio frequencies and up, aluminum and copper are better, because the field induces surface currents, reflecting rather than transmitting it. Aluminum is the metal of choice for my induction heater,
http://myweb.msoe.edu/williamstm/Images/Induction1103.jpg
because it is very effective at reflecting the 20-50kHz, high amperage fields present inside. If steel were used, it would be heated excessively by eddy currents and hysteresis loss. Aluminum runs cool.
Static magnetic fields have little effect on the average vacuum tube, after all we've been running our amps in the Earth's magnetic field for about a century with no ill effects. Use metal chassis to shield RFI, and use proper layout and wiring to exclude 60Hz.
Tim
IMO, to accomplish anything significant with iron, tin cans or mu-metal, you have to close the loop- create a full ring or better yet, a box. Just putting flat shielding between the transformer and whatever's picking up the magnetic field, does almost nothing. Using thick aluminum should be reserved for special cases when everything else has already been done and is insufficient. It helped me get the last vestiges of hum out of some extremely high conversion gain current to voltage converters in an atomic force microscope, but you can be sure there are no aluminum boat anchors in my MM phono preamp. I might consider it for MC but even there conventional techniques should work. The cheapest solution is distance.
CH
CH
I believe the thread concerned magnetic shielding only from the start.
Although the earth's magnetic field has little effect on vacuum tubes it does affect the linearity of small signal transformers such as mike-input or moving-coil cartridge step-up transformers, which is why they are shielded.
As far as using Ultraperm 80 shielding, you may need to re-anneal it after fabricating your shield. It can lose as much as 40% of its shielding ability by simply flattening it out.
John
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.