TDI Pro GB Evolution

[Carl-NC]

Is it then not possible to increase the delay on the longer delay signal as a means to both reduce the hole and achieve gb ?

All you can do is move the hole to a different tau point. You can't eliminate it.

[Reg]

Increasing the delay or time after the pulse to take the GB sample creates another problem and that is the fact that a greater gain would be required to achieve the GB. The greater gain would increase the noise you hear. Presently, most of the noise can be attributed to the increase in gain in the GB channel now.

An individual may try taking multiple samples at different times, but this may violate the ML patent which a company isn't willing to do.

As an alternative, you might simply scan the area with one GB setting and then with another but different GB setting. This moves the "hole", thus, does the same thing as taking a second sample.

Reg

[Tepco]

Some clarification may help here:
Method I described is exactly identical to GB used in GS4, only realized using different design approach more suitable for given circuit. Instead of subtracting signal from two channels, only one channel is used and GB is achieved by varying preamp gain between main and GB sample. Unfortunately, downside of this is also present, loss of depth when GB is on, and additional loss for objects close to certain TC (hole). This is inherent to two sample GB method, not to some particular detector type or design, no matter how it is realized, and there is nothing one can do about. Except changing entire design and way GB is done. Method I described cannot be implemented easily in existing detector, also it is not suitable for unipolar pulsing, (design is for bipolar) even if it can be added you wont get any improvement, just same thing made differently for particular purpose, no advantage here.

[simplejohn]

I'll tri moving the hole..... thanks Carl.

"also do you know if it is possible to separate gb samples from main signal- this way I can just add a variable delay to these samples. I might just have another look at the circuit. Then the gb can be turned right down."


I think perhaps i was not clear on the above statement.

What i meant was: if you look at the GS4 schem, Currently the signal from the pre-amp goes into both channels. what i'm proposing is leaving the signal from pre-amp to the gold channel as is, then , interupt the signal to the iron channel and add a Variable delay (not sample delay), This variable delay effectively stores the signal for 25-500 milliseconds and then releases it into the iron channel.

the result is that whenever there is a target (or a change in gb) the time difference between the 2 channels will mean a target is picked up in channel one and the subtraction deducts Ground signal from Say 100 milliseconds ago, this ground signal has had no response to the target....

you should end up with a high/low or low/high , and no hole????

this would probably produce an unstable detector though....

[Reg]

Hi SJ,

The idea you present is interesting but could get complicated. Keep in mind that the GB signal has to change with coil height, so having a proper ground balance might be impossible if the GB signal is delayed sufficiently.

I considered something similar but simply used a internally generated pseudo GB signal that would vary in amplitude with the main sample amplitude. This would work to a degree if limited to the range of the normal GB response, but I suspect it would create a hole also unless the pseudo GB signal was limited. Limiting the range of the GB signal to just sufficient GB signal would work but also affect the ability to differentiate targets also.

So, this idea does get complicated real quick but simply thinking about it as you did, does show ingenuity.

BTW, I still think there is a way to GB using a mono coil that would eliminate the hole. I just have to take time to try it. Unfortunately, the technique is quite complicated, thus isn't easy to build or try.

Reg

[Reg]

I forgot to add that if you limit the GB signal to the point that only the ground is canceled, then the hole is reduced significantly, but you sacrifice the iron reject feature.

Reg

[Reg]

SJ,

One more idea you might try and that is build a sample hold circuit and output the present GB amp to it and then back to the summing amp. In other words, insert the sample hold circuit between the GB amp and the summing amp. Now, ground balance the detector properly and have an adjustable amp that senses any increase in amplitude of the GB'd signal. If that amplitude exceeds an adjustable preset value in the sensing amp, the sample hold is initiated, thus stopping the GB amp from continuing to follow the target signal.

If working properly, the GB signal will freeze at the last ground response just before the target can influence the GB much rather than follow the target response also.

I am not sure this will work that well and reset when necessary but it should eliminate the hole effect or at least minimize it.

This concept could be modified to allow the GB signal to alter the tone but not the amplitude. Thus, the hole should be minimized while the tone won't be affected or affected much.

Again, this is just another idea that I have considered for some time but not had time to try.

Too many ideas, too little time.

Reg

[Reg]

One more thing, the TDI doesn't use a summing pot to adjust the ground balance like what is used on the GS 4. Instead, look at the hand drawn GS 5 schematic posted on this site.

[simplejohn]

Brain broken now Reg, i'll need to digest this for a bit , but seems to be good stuff if the circuit was kept simple, I'm all for simple.
A spring loaded push in GB pot could be used to innitiate the sample hold. this would minimise packaging issues...

[Orbit]

hello Reg ! whether the scheme on the forum GEOTEC goldskan 5 elemboci partially true ?

[kt315]

One more thing, the TDI doesn't use a summing pot to adjust the ground balance like what is used on the GS 4. Instead, look at the hand drawn GS 5 schematic posted on this site.

seems anything went out my zoomed eye focus... Reg, can you give direct link on the schematic? TIA

[Reg]

Look down for a post by GEO in this thread.

http://www.geotech1.com/forums/showt...ight=gold+scan

It should be post 146 on page 6.

I don't know how accurate this schematic is but the basic idea of how the two signals, the main and the GB signal, are summed is reasonably accurate and this schematic displays the difference in how the GB is done between the GS 4 and the GS 5.

Reg

[Orbit]

Ok thank you Reg!

[Orbit]

would it not be better worked lm331 in this section where are summed two channels ,I think I've tried and better worked instead of lt1014 !

[Orbit]

sorry lt 1114