Emi Frustration

[GB_Amateur]

On 8/23/2019 at 8:27 PM, GB_Amateur said:

Here's an experiment for every Equinox user.  Auto noise cancel.  Then do it again.  Do you get the same channel?

Well, I guess I should take my own recommendation.  In my backyard I turned the gain to a value that had some EMI noise, but not a crazy amount.  Then I repeated the Noise Cancel procedure for about 10 minutes.  Here are the results -- the 57 datapoints in time order and a histogram of those data:

 

 

I need to repeat this test in a different location for comparison sake, but at this point it seems that running several Noise Cancels (say 10 or so) and then choosing the most frequent (or something in the neighborhood) would make sense.  For example, here I would choose either -5 or -6.

 

 

[GB_Amateur]

I had a new experience a couple days ago that relates to this thread, so rather than to start a new one I'm just tacking it on here.

I was in one of my standard parks which is old and has produced, but grudgingly (meaning it's been searched in the past, probably at the height of the silver cleanout era when good quality discrimination detectors Hoovered up the easily recovered silver coins).  I was running the Eqx 800 with 12"x15" coil.  Near the parking lot (one of the EMI quietest parts of this often noisy park) I did both ground balance and noise cancel.  The gain was set at 22 and I was running custom 5-tones, wide open (i.e. Minelab's "all metal"), Iron Bias F2=5, recovery speed = 5.  At some point I lowered the gain to 21, then later 20, and still later back up to 21 -- I don't remember the exact gain setting I was using when the following occurred.  I was a bit surprised the detector was running so quietly at these gains, particularly with the large coil mounted.  Often at this site I have to revert to single frequencies:  10 kHz or 15 kHz being my typical retreat oases.

I got an iffy target and recalled that I had a 50 tone set mode in my Profile so switched over there to see if I cound discern more info.  The detector went crazy with a pulsing rate of 3 or 4 per second.  I looked at the mode icon and it showed Park 1.  Hmmm.  So then I did an auto noise cancel and it went quiet.  (BTW, recovery speed was also set to 5 in the Profile.)  Now that is what noise cancel is supposed to do, of course, but in my experience it seldom works so nicely.

I like to call noise cancel "frequency tweaking" because that is exactly what it is, at least on single frequency detectors.  For example, on the (simple) Tesoro Vaquero there is a central 3 position switch labeled FREQ1, FREQ2, FREQ3.  The manual tells you those frequencies:  14.3, 14.5, and 14.7 kHz.  I couldn't find the 19 frequency adjustment deltas in the Eqx manual, and maybe with multi-frequency it isn't so simple to report.

Pretty sure the source of the EMI determines if you can get a clean silencing (such as I experienced the other day) or not.  Narrow band sources (depending upon how narrow) give you a good chance of getting away from the problem by doing a noise cancel.  Broad band sources -- not gonna happen by tweaking the frequency, so try something else.

 

[Noah (FL)]

You have to noise cancel for each “program”.  
It’s not a “global” setting.  Definitely an annoying quirk but there must be a reason they have it like that????

[Chase Goldman]

To be really effective against EMI, I believe you also have to noise cancel when you go to single frequency because the mode specific Multi IQ noise cancel is N/A once you go to single.  As I've posted before, once you go to single in Park 1/2 or Field 1/2, the Multi IQ mode designation loses meaning because the Multi IQ frequency profile and processing are no longer applicable and all, you are left with are the user adjustable settings of GB, tones options, and recovery speed defaults for that mode (i.e., Park 1 = Park 2 = Field 1 = Field 2 save for the user settings at any given single frequency).

Update: I confirmed using another detector as a noise source that I could get a stable/quiet Park 1 in Multi configuration, but had to noise cancel again when I switched to Single.  In fact, as expected, every single frequency was affected differently by the noise source and each could be quieted with different Noise cancel channels.

[GB_Amateur]

There may have been some misunderstanding of what I posted.  I do realize that frequency adjust is a local, not global setting.  My point was that in some cases a change in microfrequency (via the noise cancel) can take a mode from extremely noisy to quiet.  I've never seen this before in 450+ hours of operating the Eqx 800.  Usually it's a small effect, if any.  That is, when in a quiet environment it doesn't matter what microfrequency you are in -- they are all quiet.  But in a (broad band) noisy environment it also matters little as you can't quiet things down by simply doing a noise cancel.  At least you can't with my unit -- I can't speak for others.

 

[Daniel Tn]

NASA Tom has eluded to this on his forum and I have experienced it myself with the Equinox in particular.

 

That being, just because the detector is quiet...i.e. not erratic or pulsing...does not mean that it is safe guarded from emi.  It is called silent EMI, and sometimes it is hard to pick up on it.  The Equinox seems to be more prone to it than a lot of other machines and in most of my documented cases, a depth reduction of 2-3 inches seems to be the average (compared to the same settings in a non emi area).  Tom says the Fisher CZ series is the least prone to emi of any machine he has used; but the CZs do a terrible job in the soil where I'm at.  So even in that regard, an emi hindered Equinox will do better for me than a non hindered CZ.