Jump to content

Recovery Speed Comparision


midalake

Recommended Posts

On 4/19/2019 at 2:54 PM, Flowdog said:

I am going to grab this opportunity to further explore Recovery while it is fresh and to nudge this thread a bit further hopefully. A couple of questions still dangle that came to mind while I was last in the field detecting and experimenting with recovery:

FYI,  before this particular post, I reread all of the Recovery-related links in Equinox Essentials and the cross-linked Treasure Talk authored by Steve along with all the responses on that site. Much more meaningful to me now that I have some literal skin in the game. The answers I seek may have been in this information, but I did not find them. I ask my remaining questions in an attempt to get a more nuanced understanding of Recovery without wandering into the weeds - too far. 

Does the Recovery speed adjustment set a fixed number of Tx and Rx cycles over a given period of time (something like the variable speed conveyor belt analogy Steve wrote elsewhere)? Or is the recovery cycle tied to target-dependency - suppressing sampling (cycling/responding) only once per target regardless of the recovery speed setting? 

Taking Steve's conveyor belt analogy a bit further, let's say that on that belt there were many different sized targets ranging from very tiny to very large. By running the belt at a fixed speed with a higher recovery cycle rate (6-8) does the operator give the machine a latent capacity to rapidly respond to the various target sizes only fully utilized when encountering targets in close proximity?  Or is the recovery cycle rate setting a fixed number of Tx/Rx recoveries over time, essentially forcing the machine to cycle more than once on a large target?

The reason I ask is when I recovered the biggest non-ferrous items pictured above e.g., the lead and beer can, I noted that I had only received a single dig me audio tone - just as with the smallest non-ferrous item. Size did not seem to matter to the machine in this regard.

I wonder if the machine is designed to self-limit recovery rate even at the highest settings until it processes a condition it identifies as the next target, or does it continue a forced recovery cycle rate but is designed to not introduce an identical and redundant response even while processing multiple samples within the same target?

My ears and eyes perceived a single response per target. I am wondering how Equinox pulls off this act of auditory mercy so I may better understand its language.

Thanks again Steve for all you have written on this subject. And to all those who have either authored posts or responded along the way. 

Your notion of transmit and receive cycles and recovery speed in the context of a VLF detector is a little flawed.  Don't think of it as transmit and receive cycles.  The transmit/receive process is continuous and a signal is only received (in search modes) as long as the coil is in motion.  Think of the recovery speed setting as how fast you are allowing the detector to process the signal and instead think of executing software code and microprocessor main operating loop cycles.  With modern microprocessors, processing speed is not the issue but recovery speed is also dependent on how quickly you are swinging the coil and how much target signal acquired in the process of doing so.  The detector has to process the received signal  from an individual target before it can process the next target signal, and like Steve stated in his conveyor belt analogy, until that signal is processed, the detector will not be able to detect AND process the next target until it has completed the first.  In the mean time, depending on the recovery speed chosen AND your sweep speed, you may have passed over additional targets EVEN at high recovery speeds.  So slow it down in the thick stuff.

On 4/23/2019 at 9:45 AM, glabelle said:

 This is probably pretty close. 

Remember, in the Equinox, all the analysis of a signal takes place in software. Naturally, this is Minelab's proprietary intellectual property, so we don't know exactly what they are doing. Being a retired SW engineer, I can venture a couple of guesses.

One thing they could do is spend more time, and obtain more VID accuracy when analyzing a signal when in a slow recovery mode. If you demand increased recovery speed, they might a) Spend less time in the function that analyzes the data, or b) analyze less than all 5 frequencies, or c)?

 

Just one comment here, even though ML is not revealing the secret sauce on how it transmits and processes simultaneous multi frequencies, it has been pretty well established by various third party measurements (see this thread and the links within), that Equinox is likely only simultaneously transmitting two frequencies at a time.  The frequency values used are variable based on search profile/mode settings but regardless of what simultaneous frequenceis are being used, ML is not processing each frequency separately but is instead extracting information from the received, combined frequency target signal to derive key parameters that are frequency dependent so that can see how the magnitude of those target parameters (e.g, inductance/reactance and conductance/resistivity) change with frequency and infer material type and better balance out ground effects due to mineralization and/or salinity.  This does take a lot of processing horsepower, but as I said above, that is no what is limiting the recovery speed as ML is providing recovery speed settings greater than what a detectorist can likely practically use.

  • Like 3
Link to comment
Share on other sites


6 hours ago, Chase Goldman said:

Just one comment here, even though ML is not revealing the secret sauce on how it transmits and processes simultaneous multi frequencies, it has been pretty well established by various third party measurements (see this thread and the links within), that Equinox is likely only simultaneously transmitting two frequencies at a time. 

 

With all due respect to GEOTECH (I've met him), he was measuring the waveform on the coil with an oscilloscope. That is a difficult way to determine how many frequencies are present. The best way to "see" multiple frequencies is with a spectrum analyzer. That is what I did. I used a pickup coil near the detector coil to receive whatever was being transmitted. When in "Multi", there are several (more than 2) frequencies being transmitted. I'm certain, the "secret sauce" is in how the receive signal is processed. You can see this in the diagrams below. The X axis is frequency (1KHz - 50KHz). Multiple frequencies are evident.800-600_Tx_Freq.thumb.jpg.02dd6a7dc2138acebf8c15f1996741f7.jpg

 

 

  • Like 1
Link to comment
Share on other sites

I give Geotech the benefit of the doubt that he knows how to measure the transmitted signal output of a detector since he is a successful detector designer and I have seen numerous oscilloscope and SA traces here from many folks trying to crack the transmit frequency riddle.  But yes, the "how many frequencies are transmitted by Equinox" debate will go on indefinitely as long as Minelab refuses to weigh in explicitly on the subject.  But even Minelab implies in the article linked by Steve in this post that they minimize the number of frequencies actually transmitted to the minimum needed (presumably to limited effective transmit power) so what you may be picking up are harmonics from the a fewer number of actually transmitted frequencies.  I agree the secret sauce is in the receive signal processing, but I am still pretty sure they are NOT processing each of the discrete frequencies separately and certainly not in series so the number of frequencies used/transmitted/received or whatever is probably not limiting their processing capabilities.  That is all I was saying in response to your comment that they could "analyze less than all 5 frequencies".

Link to comment
Share on other sites

I think Equinox  advantage is that it allows you to use  many power   from several "Effective" frequencies that are needed for a given detection purpose  ...
My  FFT-measurements:

 Tx Beach 1:

DS1Z_QuickPrint130.jpg

 Tx Park2 :

DS1Z_QuickPrint121.jpg

 Tx Gold 1: 

DS1Z_QuickPrint134.jpg

 

..Of course substantial work its also happens on the side of the Rx-received signal ... and its subsequent calculation -... where we can also include the use of a given level of Recovery speed ...

  • Like 1
Link to comment
Share on other sites

16 hours ago, Chase Goldman said:

Your notion of transmit and receive cycles and recovery speed in the context of a VLF detector is a little flawed.  Don't think of it as transmit and receive cycles.  The transmit/receive process is continuous and a signal is only received (in search modes) as long as the coil is in motion.  Think of the recovery speed setting as how fast you are allowing the detector to process the signal and instead think of executing software code and microprocessor main operating loop cycles

Thanks for the clarification and reiterating the "slow down in thick stuff" advice. Separating out Tx/Rx from the recovery speed adjustment setting helped me connect the multiple explanations in this thread (and other linked threads) regarding the Recovery setting. The last sentence written in the quoted response above is a bit unclear to me, so I wanted to simply say that my take away is - adjusting recovery speed alters, via software code, the time it takes to fully process a target into both TID & audio signal. At that point the processed target is "cleared" and the next detected target is similarly processed. Swing speed should allow optimum use of the recovery speed setting selected. Sine waves need not apply. ?

  • Like 1
Link to comment
Share on other sites

Highly set recovery speed also has two features:

1. "Deep" - for example, in clean terrain when setting high recovery speed 6-7-8 you get deeper signals with faster spool sweeping ... with slow coil sweeping ..you reduce effective detection depth

2. "Separation" - A high recovery setting of 6-7-8 .. allows excellent separation with normal or slow sweeping coil.
At lower recovery speed settings you can never get so good separation by normal coil sweeping .. but sometimes you can still separate well at very slow or extremely slow sweep coil of detector...

..On the other hand, the disproportionately high recovery speed 7-8 has a rather low sensitivity, but the reach of very tiny objects of chain, jewelry or nuggets  .. On objects - the size of the coin still responds very well ..- even though it is lower range ...

  • Like 2
Link to comment
Share on other sites

7 hours ago, Flowdog said:

Thanks for the clarification and reiterating the "slow down in thick stuff" advice. Separating out Tx/Rx from the recovery speed adjustment setting helped me connect the multiple explanations in this thread (and other linked threads) regarding the Recovery setting. The last sentence written in the quoted response above is a bit unclear to me, so I wanted to simply say that my take away is - adjusting recovery speed alters, via software code, the time it takes to fully process a target into both TID & audio signal. At that point the processed target is "cleared" and the next detected target is similarly processed. Swing speed should allow optimum use of the recovery speed setting selected. Sine waves need not apply. ?

It used to be that recovery speed was dependent on how fast your analog or even your digital microelectronic circuitry could process and display/sound off on the signal and as a result, detectors were limited in the maximum recovery speed available with Tesoros leading the way as they primarily focused on audio performance . Multi Frequency detectors such as the Whites V3i and the Minelab eTrac and CTX 3030 upped the ante on signal processing power needs and their focus was on better target identification at depth rather than speed.  Earlier this decade, XP focused on getting a detector out their that had many game changing attributes but the main one relevant here, was a fully adjustable range of recovery speeds (XP calls this parameter "reactivity").  While some detectors had limited ability "adjust" recovery speed by selecting specific modes that invariably compromised some performance aspect to achieve increased speed, XP was really one of the first to enable to user to change recovery speed and also make it such that the detectorist really would have to sweep ridiculously fast to challenge it at high settings.  FFWD to 2018 and Equinox came along and took a cue from XP and provided variable and fast recovery speed.

  • Like 2
Link to comment
Share on other sites

On 4/27/2019 at 11:43 PM, EL NINO77 said:

 

Highly set recovery speed also has two features:

 1. "Deep" - for example, in clean terrain when setting high recovery speed 6-7-8 you get deeper signals with faster spool sweeping ... with slow coil sweeping ..you reduce effective detection depth

2. "Separation" - A high recovery setting of 6-7-8 .. allows excellent separation with normal or slow sweeping coil.
At lower recovery speed settings you can never get so good separation by normal coil sweeping .. but sometimes you can still separate well at very slow or extremely slow sweep coil of detector...

..On the other hand, the disproportionately high recovery speed 7-8 has a rather low sensitivity, but the reach of very tiny objects of chain, jewelry or nuggets  .. On objects - the size of the coin still responds very well ..- even though it is lower range ...

Currently I am hunting in an area that requires enhancing signal separation over depth. I am curious about any noticeable reduction of iron falsing with lowered recovery speed. On the other hand, It is very helpful to know that if ever I hunt in a low mineral target-sparse site, depth may be possible at higher recovery speeds. The additional info about high recovery speed with respect to small gold nuggets could not be timed better. Thank you. Whenever I move to nugget hunting I will keep that in mind - after I get comfortable hunting in the Gold mode presets.

 

  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...