A
Anonymous
Guest
Chris,
<span style="background-color:#ffff00;">This is where we are in the topic and Glenn can make any comments he cared to to clarify. Glenn and I both know we need to be cautious about going so much into the technical side of the detector as this is of little interest to most participants. We don't want to exclude anyone but don't want to go against the purpose of the forum by getting carried away with the technical side.</span>Anyone is more than welcome to join by direct email or on the forum if that is what people want.
Sometimes the answer is not as complex as we make it and I may often be guilty of that. The sensitivity adjustment is to make the detector more sensitivity to targets. The problem is it also is more sensitivity to ground noise and other external noise. We can increase and decrease the sensitivity manually or we can have the circuits adjust the sensitivity. The reason we cannot set the sensitivity to maximum and leave it there is because of ground noise, problems with co-located targets, and the user needs to adjust the sensitivity based on preferences.
What was of particular interest to me was how a quarter responds to manual and semi-auto. If the detector is set on 1 in semi-auto a quarter is detected at 8 inches loud and clear. In manual the quarter is not detected unless the sensitivity is set to at least 16. This is for a test quarter in a specific soil matrix. If we have the sensitivity set to 28 then both manual and semi-auto enable the detector to detect the coin loud and clear.
I think the question is what are the advantages and disadvantages of semi-auto sensitivity? We know it give us a more stable threshold which means the detector is less sensitivity to external noise. We can accomplish the same this by decreasing the sensitivity manually. So, we want to know if semi-auto is "accurate" enough in how fast it makes the sensitivity adjustments and how far does it increase or decrease the sensitivity.
There appears to me to be two variable to work with and they are external noise and targets. The semi-auto means to me that the circuits are not constantly adjusting sensitivity but do so when specific parameters are met or encountered. I know that external noise is the primary problem so assume that is one parameter.
I did this to see how external noise is encounter in a couple of different areas where I search. I used the learn function and started with a black screen. I set the sensitivity and held the detector at in the air and watched as external noise dotted the screen. I did the same thing with the coil on the ground and the amount of RF induced is less. I also pointed the coil in the direction of the RF, power lines and a transformer, and the amount of induced RF greatly increased. I walked to different location at various distances to the source and could see that the amount of induced RF would vary greatly. I then checked with the coil on the ground at 0, 90, 180 and 270 degrees to the source. It was interesting to me that simply turning so the source was at my back would decrease the amount of stray induced RF both with the coil on the ground and in the air. That is the basics of my comments that the detector is very directional to RF.
This is what I concluded, that if I used manual sensitivity then it was very important to noise cancel after any change in sensitive or audio gain. In semi-auto this was less critical as the circuits compensated for the external noise. If the threshold had noise and was unstable all I had to do was push the semi-auto pad and the threshold became stable. If I went back to normal the threshold again was noisy. It is obvious that semi-auto was decreasing the sensitivity to some value that compensated for the noise. It was also obvious from my other test that if the sensitivity was set to only 1 then semi-auto would increase the sensitivity and since I could detect the test quarter at 16 in manual then I concluded it was at least to that point in semi-auto.
The question then became how did the circuits know how far to adjust the sensitivity down to decrease external noise and up to increase sensitivity to a target at an unknown depth? There are two variables, noise and good targets, that can be used or there is some kind of fixed ranges based on design. I believe it is based on a comparison of the noise to good targets and is set by Audio Gain and sensitivity. The problem with this is how does the detector establish what the faintest good target is to compare to the noise? I believe it is just a matter of where the sensitivity is set which determines the signal to noise ration. I am not sure if there are any calculation based on the good targets detected vs ground noise.
HH, Cody
<span style="background-color:#ffff00;">This is where we are in the topic and Glenn can make any comments he cared to to clarify. Glenn and I both know we need to be cautious about going so much into the technical side of the detector as this is of little interest to most participants. We don't want to exclude anyone but don't want to go against the purpose of the forum by getting carried away with the technical side.</span>Anyone is more than welcome to join by direct email or on the forum if that is what people want.
Sometimes the answer is not as complex as we make it and I may often be guilty of that. The sensitivity adjustment is to make the detector more sensitivity to targets. The problem is it also is more sensitivity to ground noise and other external noise. We can increase and decrease the sensitivity manually or we can have the circuits adjust the sensitivity. The reason we cannot set the sensitivity to maximum and leave it there is because of ground noise, problems with co-located targets, and the user needs to adjust the sensitivity based on preferences.
What was of particular interest to me was how a quarter responds to manual and semi-auto. If the detector is set on 1 in semi-auto a quarter is detected at 8 inches loud and clear. In manual the quarter is not detected unless the sensitivity is set to at least 16. This is for a test quarter in a specific soil matrix. If we have the sensitivity set to 28 then both manual and semi-auto enable the detector to detect the coin loud and clear.
I think the question is what are the advantages and disadvantages of semi-auto sensitivity? We know it give us a more stable threshold which means the detector is less sensitivity to external noise. We can accomplish the same this by decreasing the sensitivity manually. So, we want to know if semi-auto is "accurate" enough in how fast it makes the sensitivity adjustments and how far does it increase or decrease the sensitivity.
There appears to me to be two variable to work with and they are external noise and targets. The semi-auto means to me that the circuits are not constantly adjusting sensitivity but do so when specific parameters are met or encountered. I know that external noise is the primary problem so assume that is one parameter.
I did this to see how external noise is encounter in a couple of different areas where I search. I used the learn function and started with a black screen. I set the sensitivity and held the detector at in the air and watched as external noise dotted the screen. I did the same thing with the coil on the ground and the amount of RF induced is less. I also pointed the coil in the direction of the RF, power lines and a transformer, and the amount of induced RF greatly increased. I walked to different location at various distances to the source and could see that the amount of induced RF would vary greatly. I then checked with the coil on the ground at 0, 90, 180 and 270 degrees to the source. It was interesting to me that simply turning so the source was at my back would decrease the amount of stray induced RF both with the coil on the ground and in the air. That is the basics of my comments that the detector is very directional to RF.
This is what I concluded, that if I used manual sensitivity then it was very important to noise cancel after any change in sensitive or audio gain. In semi-auto this was less critical as the circuits compensated for the external noise. If the threshold had noise and was unstable all I had to do was push the semi-auto pad and the threshold became stable. If I went back to normal the threshold again was noisy. It is obvious that semi-auto was decreasing the sensitivity to some value that compensated for the noise. It was also obvious from my other test that if the sensitivity was set to only 1 then semi-auto would increase the sensitivity and since I could detect the test quarter at 16 in manual then I concluded it was at least to that point in semi-auto.
The question then became how did the circuits know how far to adjust the sensitivity down to decrease external noise and up to increase sensitivity to a target at an unknown depth? There are two variables, noise and good targets, that can be used or there is some kind of fixed ranges based on design. I believe it is based on a comparison of the noise to good targets and is set by Audio Gain and sensitivity. The problem with this is how does the detector establish what the faintest good target is to compare to the noise? I believe it is just a matter of where the sensitivity is set which determines the signal to noise ration. I am not sure if there are any calculation based on the good targets detected vs ground noise.
HH, Cody