component placement in the case

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Gurus,

Is there a general 'rule of thumb' when placing the torroid, PSU and amp boards in the same chassis when it comes to placement?

I'm wondering if this works like premise wiring - "signal cables should cross power or other sources of EMI @ 90 degree angles, and should be run 18" apart" kind of thing.

I've seen a number of designs where the PSU/Torroid is seperated by a strip of metal or in an 'internal box' of some sort.

Can anyone offer suggestions? I'm trying to keep the size of the chassis down as much as I can w/o using an external case for the PSU/Torroid combo.
 
It's only an issue when you use components with stray magnetic fields.
It's not even a problem when the magnetic field cannot induct onto something, or even inducts onto something that isn't influenced by this.

Toroids are said to have minimal stray magnetic fields, so most of the times it's not a problem...Some manufacturers place ground planes out of precaution...

I'm no expert in this field, but I was allready thinking of a simple way to measure it (iron dust on a piece of paper?)...

indeed, Guru's?
 
The large square EI frame transformer has a large but not terribly strong magnetic field coming out the curved "bell" ends of the transformer in broad circles with the transformer windings (under the "bells" that cover the windings) acting as the anchor point. There is little or no field radiated by the fat metal laminations. You do need to be careful where the magnetic field of the EI frame transformer is aimed. You want to keep it away from the low-level circuit boards and from the input RCAs and wiring.

The toroid, a round donut-looking transformer, has almost no magnetic field radiating out from it except along an almost straight line right through the hole in the center of the toroid. Within that center hole and extending up and down (toroids are almost always mounted flat because of this magnetic field) from it is one massively strong magnetic field. It has the same relative strength as the weaker but larger-sized field that radiates from the EI frame transformer, but it is so concentrated by the shape of the toroid transformer that that the density of the flux energy in the field radiating from a toroid is very high indeed. The larger the toroid, the stronger this radiated field. Imagine a miniature javelin going through the hole in the toroid transformer; this is a pretty good representation of the radiation pattern of the magnetic field (though the field spreads out a bit as you get farther away from the toroid). Placing some other component above or below an amplifier with a large toroidal transformer inside is asking for trouble. You could get hum or otherwise degrade the sound. Only a large amount of ferrous material, like cast iron, above and below a toroidal power transformer will block the magnetic field. The aluminum or thin steel chassis in most audio components is virtually transparent to the magnetic field from the transformer. So are wood or glass rack shelves. Even the second chassis of another nearby component won’t reduce the strength of the magnetic field too much. Other than an inch-thick cast iron plate, the best defense against these strong magnetic fields from toroidal transformers is distance or placing other components beside the component with the toroidal power transformer rather than above or below it.

Even the small toroid in a turntable or preamp power supply could induce hum in other components if they get to close to the javelin-like magnetic field radiating out from the hole in the center of the toroid.

Doug Blackburn


http://www.soundstage.com/maxdb/maxdb091998.htm
 
Nordic said:


Doug Blackburn
...

Something doesn't seem right about this stuff about the magnetic field along the axis of a toroid. Magnetic flux is perpendicular to the current through the coils that produce it (the good old right hand rule). That would do exactly what we expect and send the flux through the loop of the core.

There will be some axial flux because the coils of wire that wrap around the core essentially form a "super" coil of the same loop size as the core of the transformer (assuming they go all the way around the core radially), but they form only a 1 or 2 (maybe 3) turn loop. The permeability of the air is small compared to the core, and the opening is relatively large, so I'm pretty sure you won't find a "strong" field axial to the transformer core.

If someone has an EM field solver or a gauss meter we can find out about this real fast.

Anyway, it's best to keep low level signals and high impedance circuits away from high current parts like power transformers, so the transformer usually goes at one end of your box and the circuits at the other.

I_F
 
Keyoke said:
I've seen a number of designs where the PSU/Torroid is seperated by a strip of metal or in an 'internal box' of some sort.

falcott said:

If one is forced into positioning the transformer close to the amp, maybe he can use some steel shielding between them? If so, does the thickness of the steel make any difference?

Any thoughts on how effective the above would be? I am planning to implement something similar to be above for my PSU. I may have situations where the DC power wires run along the same line/space with signal. Planning to separate the DC wires with a thin steel plate.

Any opinions?

Thanks
 
v-bro said:
By the way, I just took a large air-core speaker filter coil and set my DMM on Ac mV, if I hold it near a PSU tranny it measuers a couple of mV, there's a simple Gauss meter!

This serves to illustrate my point about keeping high impedance circuits away from transformers, but it isn't anything like a Gauss meter. A Gauss meter measures magnetic flux. The flux you are trying to measure is proportional to the current in the transformer. If you want to measure the flux, you can do so by reading the current induced in the coil. The difference between reading the voltage and the current is that a volt meter has very high resistance and an amp meter has a very low resistance. (In a DMM, the amp meter mode is actually a very low resistance shunt with the very high resistance volt meter reading the voltage drop across the shunt).

Try your experiment again with the coil connected to the current input on your meter.

I_F
 
I think it can be pretty effective, but don't know details on how thick it would have to be... I think it depends on the flux and saturation level, probably 1 or 2 millimeters thickness will handle it well enough in this case...

But to reduce size it would probably not matter much......a 'box in a box' is still practically 'space in space'....;)

Thanks for pointing out what I forgot I_forgot....
On 'mA' setting it reveals the size of the magnetic field quite well...
 
I_Forgot said:


So is it gargantuan or merely stupendous?

I_F


It measures about 5mA from a 2mH aircoil placed on top of my Weller WTCP-S. Measured with Fluke 179 DMM on mA(ac)/Hz setting. Nice to see is the rate fluctuates as soon as the thermostat kicks in....

I think with a coil with thinner wire and larger mH (like a relay coil) can make it a lot more sensitive...
 
This is the setup:

I will try it soon with other coils and see what happens...
 

Attachments

  • gaussmeter.jpg
    gaussmeter.jpg
    41.1 KB · Views: 160
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.