EMI shield

[Ferric Toes]

I have got confuses

are we talking about EMI shielding of the electronics

or the coil shield ?

You are correct in your post 14 that "most of the EMI is picked up by the coil", so that is where the discussion should centre. There appears to be a bit of confusion is some quarters about EMI which stands for 'ElectroMagnetic Interference' from an outside source. The radiated signal from the coil is not EMI, unless of course it interferes with other electronic apparatus, or even another detector. Shielding the coil has another beneficial effect in that it prevents coil to ground spurious responses caused by wet grass, wet ground, seaweed, wet beach, hand capacity, etc. The coil sees only the capacitance to the shield and not to exterior things. Personally, I would never use an unshielded coil for any application.

Eric.

[green]

I have got confuses

are we talking about EMI shielding of the electronics

or the coil shield ?

Was trying to understand shielding. EMI shielding of electronics or coil, pick either one or both. EMI pickup from the coil is my main problem. Shielding the coil with graphite paint solves the problem getting a signal when my hand is brought near the coil. Don't think it reduces EMI pickup.

[green]

This is the tape I use very successfully. I buy it from a company called 'cidimall' on eBay. Woven cloth is nickel plated copper and another good feature is that it is backed with conductive adhesive. I wrap the bare coil with PVC spiral cable wrap and then helically wrap the tape on top of this which gives mm or so spacing. For the last few turns of shielding, I wrap it over a couple of inches of bared 7/0.2 stranded wire which connects to the shielding via the conductive adhesive. the tail of the wire, with a bit of the retained original insulation, is then connected to coax ground/coil end ground. There must be a small gap of course to prevent a shorted turn effect. Wrap around shielding is by far the best, because coaxial construction around the coil bundle results in minimal eddy currents in the shield and least effect on the TX waveform.

Eric.

Have you compared EMI pickup from the coil, shielded and not shielded looking at amplifier out. https://www.geotech1.com/forums/atta...2&d=1552349592 My try. Should I be trying a different test method? How close to shorted input should shielding get?

[Tinkerer]

Was trying to understand shielding. EMI shielding of electronics or coil, pick either one or both. EMI pickup from the coil is my main problem. Shielding the coil with graphite paint solves the problem getting a signal when my hand is brought near the coil. Don't think it reduces EMI pickup.

The human body seems to be a good antenna. When I made an FFT of the EMI picked up by the coil, from my hand it looked like the EMI from the computer and light, amplified.
For the shielding of the coil, I think we need to look at the frequency of the EMI. We want to pick up the response from the targets. A target with a TC of 1us, corresponds to about 160kHz. So we should keep the shielding passband below that frequency.

However, the TX pulse has some high frequency components. These might get attenuated by the shielding.

Then there is the TX pulse decay. If the shielding stop band is too low, the decay will be slowed down and a longer sample delay results.

Shielding the cable:
A Co-ax cable is already shielded. Shielding gets more complicated when there are several wires in the cable, like for a DD coil or bi-polar TX with a concentric balanced receive coil and a bucking coil.

[Ferric Toes]

Have you compared EMI pickup from the coil, shielded and not shielded looking at amplifier out. https://www.geotech1.com/forums/atta...2&d=1552349592 My try. Should I be trying a different test method? How close to shorted input should shielding get?

Your test method looks good. I get similar results and have to switch off my desk lamp and solder station to calm things down. Aligning the angle of the coil for a noise null also helps as there are outside sources of noise that I haven't been able to trace. I am looking forward to my move to a new workshop as the ambient noise level there is much much lower. Is it correct that you have a noise cancelling figure 8 RX coil within a mono TX? Noise cancelling coils are good for low frequencies, but the halves need to be carefully balanced to achieve useful rejection above 50kHz. On the rare occasions that I have used this type of coil, I also shield it with the EMI tape as well (a tedious job).

I don't think you will ever get close to shorted input noise, as the shielding must not restrict the bandwith that the detector needs to accurately pass the decay curve. Tinkerer raises this important point. It is amazing how much rf noise exists below 200kHz, including a French broadcast station that gives a strong signal here in UK.

Eric.

[green]

Your test method looks good. I get similar results and have to switch off my desk lamp and solder station to calm things down. Aligning the angle of the coil for a noise null also helps as there are outside sources of noise that I haven't been able to trace. I am looking forward to my move to a new workshop as the ambient noise level there is much much lower. Is it correct that you have a noise cancelling figure 8 RX coil within a mono TX? Noise cancelling coils are good for low frequencies, but the halves need to be carefully balanced to achieve useful rejection above 50kHz. On the rare occasions that I have used this type of coil, I also shield it with the EMI tape as well (a tedious job).

I don't think you will ever get close to shorted input noise, as the shielding must not restrict the bandwith that the detector needs to accurately pass the decay curve. Tinkerer raises this important point. It is amazing how much rf noise exists below 200kHz, including a French broadcast station that gives a strong signal here in UK.

Eric.

Yes. I use a figure eight receive on my PI, for me less noise than a mono and it does reduce ground signal. I have a 1.5inch figure eight Rx and a 8inch figure eight Rx for my Target Response Tester. The eight inch has more noise. Noise is a problem when charting log out. Log out is full scale minus when log in equals minus noise. Been wondering if I could EMI shield the eight inch.

[Orbit]

French broadcast station burn ...

[kt315]

the Nature (Mother) Eric of EMI is working/effected in both sides. you protect a coil from EMI and transmitter EMI force/energy
from a coil also falls down. received EMI also falls down from a target - you catastrofically lose sensitivity. so all you try to do is only absurd idea.
there is just one way to work in EMI noise conditions - to do simple VLF project - Fisher 1210, Tesoro, Bounty Hunter IV, etc. detectors with low sensitivity.

[Ferric Toes]

I took a piece of my 25.4mm (1") wide nickel/copper woven tape and measured it's resistance. I placed the probes 1" apart and read 0.1ohms, which is the smallest value my multimeter will resolve. This is not the most accurate way of measuring the resistance of what is a thin sheet (0.09mm) but it is not far out. By comparison the graphite coating on the inner surface of a couple of Whites shells, similarly measured was 50 - 100 ohms. This coating will obviously be a less effective EMI shield than the nickel/copper tape.

A square piece of the tape 1" x 1" gives no reaction when place on a solenoid coil with a sample delay of 10uS. This is the best position to generate eddy currents in the material as the field from the coil will be perpendicular to the surface. It takes three pieces of that size, stuck together with the conductive adhesive, to get even the slightest reaction. The single layer helically wound on the cable wrap holding the coil will not have eddy currents generated, even if you have two or even three layers of this tape, for the reason that the collapsing field is concentric with the helical tape and at no time is perpendicular to it's surface. From a manufacturing point of view, it is very time consuming to apply this form of shielding, but for small numbers or experimenters, it is the best form of shielding; in my view anyway.

Eric.

[Qiaozhi]

This is the best position to generate eddy currents in the material as the field from the coil will be perpendicular to the surface. It takes three pieces of that size, stuck together with the conductive adhesive, to get even the slightest reaction. The single layer helically wound on the cable wrap holding the coil will not have eddy currents generated, even if you have two or even three layers of this tape, for the reason that the collapsing field is concentric with the helical tape and at no time is perpendicular to it's surface. From a manufacturing point of view, it is very time consuming to apply this form of shielding, but for small numbers or experimenters, it is the best form of shielding; in my view anyway.

Eric.

Eric - When you wind the tape around the coil, are you overlapping the turns or leaving a small gap between each turn?

[Ferric Toes]

the Nature (Mother) Eric of EMI is working/effected in both sides. you protect a coil from EMI and transmitter EMI force/energy
from a coil also falls down. received EMI also falls down from a target - you catastrofically lose sensitivity. so all you try to do is only absurd idea.
there is just one way to work in EMI noise conditions - to do simple VLF project - Fisher 1210, Tesoro, Bounty Hunter IV, etc. detectors with low sensitivity.

That is a strange idea that EMI falls down from a target. What is happening in a PI is that the pulse field generates eddy currents in the target which are sychronous with the transmitted pulses; hardly EMI as that is what you want to happen. This sychronism is in fact used in the sampling and integrating parts of the circuit to extract the wanted signal from non-synchronous noise (EMI). EMI can be reduced still further by integrating over a longer period or increasing the pulse repetition rate. There has to be a compromise at some point though, so as not to slow down the target speed response of the detector, so the best way to tackle EMI, and the first line of defence, is to shield the coil. Done properly, this can only improve the noise/signal ratio and improve the sensitivity.

Eric.

[Ferric Toes]

Eric - When you wind the tape around the coil, are you overlapping the turns or leaving a small gap between each turn?

I overlap; not fully but by about 5mm using the 25mm wide tape. For smaller coils i.e. 10" or less, I use a narrower tape such as 20mm which gives a neater finish. Some minor rucking up is inevitable, but is of no consequence.

Eric.

[Tinkerer]

https://www.aliexpress.com/item/Radi...27424c4djUoSxO

Here is another product for shielding, worth trying. It is silver coated Nylon fabric.

[KingJL]

I overlap; not fully but by about 5mm using the 25mm wide tape. For smaller coils i.e. 10" or less, I use a narrower tape such as 20mm which gives a neater finish. Some minor rucking up is inevitable, but is of no consequence.

Eric.

I also use this tape for shielding. I have applied it overlapped, as Eric has described, and non-overlapped with about 2mm (actually varies from ~2mm to ~4mm) spacing. Both seem to perform equally. I have also used it for shielding my probes. Note however, do not overlap when shielding probes... you will not like the result!

[Qiaozhi]

I overlap; not fully but by about 5mm using the 25mm wide tape. For smaller coils i.e. 10" or less, I use a narrower tape such as 20mm which gives a neater finish. Some minor rucking up is inevitable, but is of no consequence.

Eric.

Since you can have gaps in a Faraday shield, as long as any holes are significantly smaller than the wavelength of the radiation, I was just wondering how wide the gap could be before the shield becomes ineffective. I've always thought of the coil shield as being used primarily as an electrostatic shield, rather than to block EMI. Personally I've tended to use 20um aluminium tape with a slight overlap, which was available from Maplins before they went bust. Not long ago, however, I did buy some of the same tape you're using from eBay, but haven't had the chance to try it out yet.