Emi. Oh My!

[Jeff McClendon]

By EMI noise reduction in this case, I am assuming you mean shielding, filtering, software factors and factory chosen lower gain in general that can reduce overall EMI susceptibility BEFORE doing a noise cancel procedure if the detector being used has one. Sensitivity to smaller, hard to hit targets might be affected depending on what frequency/frequencies are being used.

Just my opinion based on using a lot of different detectors in urban areas.

[Past Member]

So why is it that in my high emi sites, the noise reduction on my Vanquish 540 and my Legend do nothing? Why is it that of the countless videos I've seen of those detectors, or the D2, Nox, etc, the noise pretty much stays the same after noise cancelling? I'm thinking noise cancelling only works under a rare emi situation in which the emi is only at one particular frequency (or something along those lines).

Well, what I have been doing at each high emi site, is burying my desired target very deep, and seeing what hits it best. Be it SMF with reduced sensitivity, or SF with maximum or close to maximum sensitivity.

[midalake]

2 hours ago, Digalicious said:

So, would it be a true statement to say, "The more EMI noise reduction, the less sensitivity"?

No, incorrect. 

[jasong]

Noise cancellation isn't really just 1 concise thing. EMI can be narrow or wide band in nature, and that changes how it's dealt with.

A lot (if not most) of bad EMI is wideband impulses (take lightning for example) - meaning they encompass many frequency bands simultaneously. But some EMI is very close to 1 single frequency - think of a radio transmitter broadcasting on 1 channel only. How close any given EMI source is in frequency to the detector's various frequencies you are using will determine how much noise you hear. Wideband sources are generally much harder to filter out than single frequency sources. 

You can tune out the single frequency noise by channel hopping (slight changes in the primary detector frequency). This does not result in sensitivity loss. Wideband EMI requires more sophisticated measures to cancel the noise out, usually involving hardware filters and software algorithmic processing. These may result in minor sensitivity loss by proxy of audio processing removing parts of the audio data stream (Smoothing or the Stabilizer are two examples of this on a ML PI detector). I don't know how/if wideband cancellation is done on multifrequency VLF's so I have no comment on them. 

Decreasing sensitivity sometimes reduces noise because what you are doing is reducing RX amplification. That means you aren't amplifying the EMI as much. It also means you aren't amplifying the target signal as much, and that's why this method results in less sensitivity, while the channel hopping doesn't. 

You can also try to cancel noise by physically shielding components - the coil, cable, and control unit are all susceptible. These approaches do not result in sensitivity losses either. In fact, they might result in sensitivity gains since good shielding allows you to run higher gains in high EMI areas as well as hear much more subtle signals. Which makes the absense of robust shielding on the world's most expensive detectors recently all the more confusing to me, but that's another story altogether. 

[Past Member]

7 hours ago, jasong said:

Noise cancellation isn't really just 1 concise thing. EMI can be narrow or wide band in nature, and that changes how it's dealt with.

A lot (if not most) of bad EMI is wideband impulses (take lightning for example) - meaning they encompass many frequency bands simultaneously. But some EMI is very close to 1 single frequency - think of a radio transmitter broadcasting on 1 channel only. How close any given EMI source is in frequency to the detector's various frequencies you are using will determine how much noise you hear. Wideband sources are generally much harder to filter out than single frequency sources. 

 

 

 

So I guess I was on the right track when I mentioned that: I'm thinking SMF noise cancelling only works under a rare emi situation in which the emi is only at one particular frequency (or something along those lines)

That would also explain why the emi noise reduction feature in SMF mode, typically has little to no effect.

Regarding coil shielding:

To block noise, doesn't the shielding have to be made of metal? If so, wouldn't that seriously decrease performance? Or, can coil shielding be a non-metalic shield on top of the coil that doesn't affect the field being emitted from the bottom of the coil?

[JCR]

Any shielding would move With the coil, so it is effectively Not seen by a Motion based detector.

EMI mitigation has always seemed to me to be a large area of potential performance improvement. The reports on the Manticore show some significant advancement over the Equinox 600/800 in this regard. I know my Legend does a much better job of handling EMI than my Anfibio Multi. So I hope for further improvements in this area. Manufacturers seem to be aware of the need for improvement.

Then, you have Silent EMI......

[Jeff McClendon]

2 hours ago, Digalicious said:

That would also explain why the emi noise reduction feature in SMF mode, typically has little to no effect.

I have used an Equinox 800 for over 5,000 hours estimating the times from my hunt logs, over 700 on the Legend and approaching 500 for Deus 2. Most of that detecting was done in two different urban/suburban areas that both have a population of over 2 million humans. I have never had to pack up my toys and go home using these detectors due to impossible to mitigate EMI, not even close. Using several hot single frequency detectors that people still rave about......I absolutely had to put those back in the car at some on the exact same sites where the SMF detectors above could still detect very well in one of their SMF modes. I have had two instances dealing with cattle fences, one instance with an electric pet fence and one with overhead power lines where I had to switch to a single frequency on the Equinox. 

So, my experience is that EMI is reduced enough using these detector's noise cancel/reduction features to get the job done.

Now, if I try to use one of these detectors in my house or in my backyard for testing......that is a whole different story which relates to some of your references to testing videos made by YouTubers on back porches, front porches, concrete slabs, backyard test gardens in urban/suburban areas, etc. There is no way to completely cancel out EMI from millions of wireless communication systems and all the power lines  in these areas using pulse induction or hot SMF VLF detectors.

[jasong]

2 hours ago, Digalicious said:

So I guess I was on the right track when I mentioned that: I'm thinking SMF noise cancelling only works under a rare emi situation in which the emi is only at one particular frequency (or something along those lines)

That would also explain why the emi noise reduction feature in SMF mode, typically has little to no effect.

Regarding coil shielding:

To block noise, doesn't the shielding have to be made of metal? If so, wouldn't that seriously decrease performance? Or, can coil shielding be a non-metalic shield on top of the coil that doesn't affect the field being emitted from the bottom of the coil?

Coil shielding doesn't need to be metal per say, it just needs to be conductive. On PI detectors the coils usually have a conductive (I believe it's graphite or carbon based) paper or paint on the top/sides of the shell. 

But yeah as JCR mentioned, it moves with the coil. So no interference.

Another source of noise - as I discovered with the 6000 - is actually re-radiated EMI too. Like Jeff, I noticed that electric fences (common with ranchers where I go) were interfering. But then I noticed that normal barb wire fences were interfering in some places too, from a good distance away. There are fences everywhere around me in some places. What's happening I'm sure is re-radiation of EMI as they act like giant mile long antennas.

I also found an area about 20x20ft square fenced in. Figured I'd detect inside to see why the heck this square was fenced in the middle of nowhere. My 6000 was completely unusable anywhere inside the square - even with coil at chest level. Why?

Because the detector was interfering with itself! There is no cancelling that out. The fence wire was a large loop an inducing and re-radiating the detector's own signal. So in some cases around long conductors or large wire loops, the detector itself can be part of the EMI, which I mention here just because it never occurred to me before that trip. And if it happens - no amount of pressing the noise cancel button will ever help.

[Jeff McClendon]

19 minutes ago, jasong said:

Because the detector was interfering with itself! There is no cancelling that out. The fence wire was a large loop an inducing and re-radiating the detector's own signal. So in some cases around long conductors or large wire loops, the detector itself can be part of the EMI, which I mention here just because it never occurred to me before that trip. And if it happens - no amount of pressing the noise cancel button will ever help.

Yet another example of why expecting anything good to happen near or in a structure full of metal and electricity (house, building, car) when trying to noise cancel a PI or hot SMF detector is asking for the impossible at this point.