[Boatanchors] Need magnetic shield for shaded pole motor

[Brian Clarke]

Hello Jim,

Thank you for those kind words and compliments.

Just a couple of points on your ideas: there is no such thing as a single magnetic pole; for 60 Hz attenuation, ceramic ferrites will be very large. Super-conducting ceramics will only have a very limited effect on H-field emanations.

There are some simple rules for handling EM interference of the 60 Hz (and harmonics) variety: 
1. attenuate the emissions at the source. (as you have been trying)
2. if you can't follow rule 1, then attenuate the emissions received by the 'victim'.

I'll expand. 

1. Source
A single, no leaks iron shield around a source of EM radiation can at best achieve 40 dB attenuation. At that point many people give up and say, "Too hard." But if you go the extra mile and nest another shield around the first, you can achieve 120 dB attenuation. Attenuating emissions at the source is the preferred solution, because it means you are not producing emissions that can upset other, unknowing victims, eg, CRTs. However, in amateur radio applications, you may end up attenuating 60 Hz radiation from circuits that don't care, eg, RF circuits.

2. Victim
There are two things you can do here:
a. shield the victim; same ideas as for the source - try one, fully-enclosing, no line-of-sight holes shield; if that is not effective enough, go for two nesting shields.
b. avoid ground loops; below 50 kHz, ground only one end of any signal-carrying shielded lead, usually the victim end. Then return all power and signal grounds to a 'star' point - ensure no single lead carries both power and signal. The best way to avoid ground loops is to do what the professionals do - daisy-chain AF circuits with balanced, floating signal leads. Now, that's easier said than done, because we amateurs are so used to single-ended, unbalanced, non-floating AF circuits. One easy way to achieve balanced and floating signal leads is via transformers. Now, for the audio range amateurs frequent, ie, 300 to 3,000 Hz, there are plenty of tiny transformers going into landfill on discarded telecoms routers and switches - even new, they can be picked up for around $2 each. However, transformers are susceptible to 60 Hz mag fields - so, you may need to shield your transformers. You may object that the transformer is designed to transfer 300 to 3,000 Hz signals, but the core can become saturated by 60 Hz emissions and may then distort the signal you want to transfer because the core is working way off its approximately linear B-H curve; feeding too much DC into the primary will have the same effect. If using transformers is not your bag, another way is to use four single-ended ICs, eg, 741s, or two double ICs, eg, 747s, or one quad IC, eg, a 324. Connect two of them to produce a split-phase signal - both parts 180 degree apart; then feed these out-of-phase outputs into the other two ICs wired as unity-gain amplifiers. This kind of approach was well documented in the National Semiconductor and RCA Application notes of the 1970s. If you can't access these, give me a hoy and I'll try to scan some to you.  

73 de Brian, VK2GCE.
  On: Saturday, January 04, 2014 2:44 PM, Jim said:


  Good evening Brian etal,

  Haven’t heard from you for a while.

  The problem is with the design of a shaded pole motor which is an incredible Gaussian mess by nature.  The problem is with the 60 cycle H field that surrounds the motor which can be felt with an E core up to 6 inches away.  The aluminum chassis and the use of a copper shield is essentially invisible to the H field and does nothing to contain it.  

  The mu metal or high permeability metals have been used to contain an H field but with limitations.  It can be used very successfully as a magnetic short circuit.  If you expose a sheet of mu metal to a north or south pole individually, you will find it on the other side with little attenuation.  The point is, mu metal or high permeability metals are great to short circuit a north and south pole but have little effect on a single pole.

  The only way that I have seen non magnetic materials do low frequency H shielding is with super conductive ceramics.

  I have pretty much solved the problem using a simplistic and perhaps silly method.  I turned the motor so the coil is toward the top cover.  This moved it away from the audio preamp tube and circuits.  About a 70% fix.  Next I coupled the H field through an E core to the area of the audio tube and circuits finding a path that coupled an out of phase field to the effected area.  Voila, now 100% all is well.

  Now I can tune a –140dbm 90% modulated AM signal with a 3db S/S+N ratio, distinguish the sidebands, and not detect hum.  I am finished with this thing.

  Your have been very generous with your ideas and council and enjoy the friendship of the list.          

  Kindest regards Jim K9AXN