Optimal PI timings

[dantech]

Hi,
Could anyone explain what are the best timing for PI? I mean pulse and sample widths and delays.
I don't know if when I change delay between pulse and sample 1 I should shift sample1 together with sample2. It appears that sample 2 shouls stay in place whilst sample 1 should be shifted. Is it true?

THX in advance

Pawol

[1843]

Hi Pawol,

No!
Sample2 occurs a few hundred microseconds after the sample1 (Not after Tx pulse). and by changing the sample1, sample2 should be changed too.

Regards,
1843

[dantech]

Hi,
I know basis of PI. When I change delay1, should sample2 be also shifted in time. It appears that sample2 should stay in the same place but depending on delay 1 , the sample1 shifts of course. Generally I see when I change delay 1 and shift sample1 in time then sample 2 is shifted by the same time.
Here my question.

Secon question is what are the best timings. I have noticed when I use 40us samples width then PI is not sensitive for gold. When I use 10 us one is very sensitive. Probably the falling slope for gold ends suddenly whilst, let's say, 15us (delay1) + 40 us(sample1 width). It appears that 40 us widt for samples is too long when we want to detect gold.

[Ferric Toes]

Hi Pawol,

Generally I make the sample pulse width the same as the delay. e.g. if the delay is 15uS then that is also the width of the two sample pulses. For small gold nuggets and gold jewellery the second sample pulse can be closer to give better low frequency noise cancellation. In the Goldquest SS, the shortest delay is 10uS, then a 10uS sample pulse, then 40uS from end of 1st sample pulse to the start of the second 10uS sample pulse. Just right for nuggets and gold rings. Your 40uS sample pulses will reduce sensitivity to these items but be OK for objects with longer decays i.e. copper coins, dimes, quarters, silver coins and larger gold. Also for the latter targets you need at least 100uS between the two samples.

Eric.

[Geo]

Hi. If Eric talk what we can say
Hi Eric. I found that for coins and bigger objects is needing a 40 us delay at least. I have modified the XR71 (wood-tech) for 20us delay but as i saw don't need it
Yor comments ???
My Best Regards

[1843]

Deltapulse at delays lower than 35us becomes unstable...

[Geo]

Deltapulse at delays lower than 35us becomes unstable...

No. I have tried it at 15us without problem. Except if your coil have more capacitance and can't "follow" so thin pulse.

[Max]

Hi. If Eric talk what we can say
Hi Eric. I found that for coins and bigger objects is needing a 40 us delay at least. I have modified the XR71 (wood-tech) for 20us delay but as i saw don't need it
Yor comments ???
My Best Regards

Hi Geo,
my friend... 40uS ??? Seems a big value to me ! I've found coin sized object and coins on beach with 15uS setting at 1st delay. To me is very strange that you need 40uS with your delta or xr to do the same.

I've done lot of work on pulse detectors... many I've built myself... some I've also designed by my hands... and I can swear that 15uS is a very good timing for first delay in a good number of conditions...

But maybe you mean at "particular" inland sites conditions... and if so... I can agree with you that to avoid "some" problems with kind of soil etc etc is better pushing up first delay... and losing the small gold items if any.

Best regards,
Max

[Delbert grady]

frequency

so what about pulse rate then 10hz 100hz 200hz ???????

[Geo]

Hi Geo,
my friend... 40uS ??? Seems a big value to me ! I've found coin sized object and coins on beach with 15uS setting at 1st delay. To me is very strange that you need 40uS with your delta or xr to do the same.

I've done lot of work on pulse detectors... many I've built myself... some I've also designed by my hands... and I can swear that 15uS is a very good timing for first delay in a good number of conditions...

But maybe you mean at "particular" inland sites conditions... and if so... I can agree with you that to avoid "some" problems with kind of soil etc etc is better pushing up first delay... and losing the small gold items if any.

Best regards,
Max

One hour before leave for Ivconic
Yesterday i made some tests. The thinks are not so easy. I have a old bronze???? coin that it is detected better at 40us,
Other 2 bronze coins are detected better at 25us.
Now the strange one gold sterline (you know Max)is detected better at 30us... 30+us and a very old iron horse petal at 55us.
So when we are going for detecting where we must regulate the first Delay?
I want to detect everything under the ground
Regards

[Max]

so what about pulse rate then 10hz 100hz 200hz ???????

Hi Delbert,
I swear that frequency have importance only when you pass 1KHz so from e.g. 100Hz vs 3KHz you'll notice very well difference: 3KHz will be more sensitive to "small" stuff ... rings and the like.

At lower freq. like 50, 100 or 400Hz what change that you can notice is response time of unit: I've made tests on deltapulse (and other stuff too) and find a good setting at 450Hz... but that happens cause of integrators action... the more the pulserate the more speedy will be response of MD to targets.

In e.g. depth PI is easy finding freq. of e.g. 100Hz cause coils are great and objects are at depth... so moving the big 1m*1m coil is not like moving a small e.g. 20cm coil: you move big coils slower than smaller coils.

With smaller coils (for smaller objects at few depth) you need a very fast thing... like some Eric Foster's designs (e.g. the pulsestarII at 3KHz).. where with a DeltaPulse with 1m*1m you can reduce freq. to 100Hz or less..cause you move slower...so integrators can integrate at a lower rate without you lose targets.

That's all about frequency in PIs! It's not about skin-effect like in VLFs but about integrators time constants and pulserate required for good operations at various conditions.

Analog designs are often not so bright cause of that things...constants I mean... but a variable freq. control could solve many problems, and also about interferences , you find spots where "there are less" (positions where intermodulation with e.g. 50 or 60Hz is lower)... like in Eric's machines (pulsestarII again as example).

Kind regards,
Max

[Max]

One hour before leave for Ivconic
Yesterday i made some tests. The thinks are not so easy. I have a old bronze???? coin that it is detected better at 40us,
Other 2 bronze coins are detected better at 25us.
Now the strange one gold sterline (you know Max)is detected better at 30us... 30+us and a very old iron horse petal at 55us.
So when we are going for detecting where we must regulate the first Delay?
I want to detect everything under the ground
Regards

Hi Geo,
I found that on beach is better using 15uS (but sometimes also 10uS are very good... but you need fast coil etc to go so down on sampling).

On inland sites I've found that the 15us is, generally speaking, crap.

Yes... you can use 15uS too on inland places... but man... you have to be brave.

I found that good positions starts at 22uS to 35uS for small items using circuits like delta running at 450Hz. But sens to gold is not that good as with 15us at 22 or (much worse) 35us.

Inland sites have too different stuff on soil to be good searching there at 15uS that's the reason !

On workbench I detect small gold (and all the rest) at very deep with minimum delay... but then I lost on troubles on inland soil: ceramics, rocks, whatever make impossible using so small delay there in 95% of my places....so it's totally unuseful that way. Too false signals.

I was surprised from your 40us cause I never need of reaching that timing to gain good stability on inland sites. But then problems are never solved with sensitive PIs and I cannot cancel out some bad influence from soil... so then tired of all that story I often switch to VLFs.

Less depth but much more "relaxing".

Best regards,
Max

[Delbert grady]

Thank You Max

[Geo]

So a VLF with a 15....18 in homemade and not heavy coil is the best way.
Except if we must use big coil 1x1 or 2x2 m. And if we are looking for big quantity of gold with a coil 1x1m who must be the delay????
If the gold is inside mettalic box (from iron or Alum)?????
(15 or 35...45us???)
Regards

[Ferric Toes]

Hi Delbert,
I swear that frequency have importance only when you pass 1KHz so from e.g. 100Hz vs 3KHz you'll notice very well difference: 3KHz will be more sensitive to "small" stuff ... rings and the like.

At lower freq. like 50, 100 or 400Hz what change that you can notice is response time of unit: I've made tests on deltapulse (and other stuff too) and find a good setting at 450Hz... but that happens cause of integrators action... the more the pulserate the more speedy will be response of MD to targets.

In e.g. depth PI is easy finding freq. of e.g. 100Hz cause coils are great and objects are at depth... so moving the big 1m*1m coil is not like moving a small e.g. 20cm coil: you move big coils slower than smaller coils.

With smaller coils (for smaller objects at few depth) you need a very fast thing... like some Eric Foster's designs (e.g. the pulsestarII at 3KHz).. where with a DeltaPulse with 1m*1m you can reduce freq. to 100Hz or less..cause you move slower...so integrators can integrate at a lower rate without you lose targets.

That's all about frequency in PIs! It's not about skin-effect like in VLFs but about integrators time constants and pulserate required for good operations at various conditions.

Analog designs are often not so bright cause of that things...constants I mean... but a variable freq. control could solve many problems, and also about interferences , you find spots where "there are less" (positions where intermodulation with e.g. 50 or 60Hz is lower)... like in Eric's machines (pulsestarII again as example).

Kind regards,
Max

Pulsestar isn't my design, but Deepstar is and runs at 3kHz. PulseStar is TB Electronics in Germany. My Goldquest SS runs at 10kHz pulse rate and 10uS delay and sampling. This is very good for small targets and fast response, but 10uS delay is about as fast as you can go on a saltwater beach. The fastest PI I have made runs at 100kHz and 1uS delay, 1uS samples. It has industrial uses for finding broken off needle tips in fabric, but you could not use it for ground searching. Some PI mine detectors run at a few hundred Hz and are still sensitive to small metal, so pulse frequency is of rather less importance than fast switch off of the TX and short delay. High pulse rates are better for integrating out random noise and low frequency noise. Detectors with low repetition rates have more problems with power line noise and its harmonics. A frequency shift control is always a good idea to prevent the beat note effect when interference and the pulse rate are near sychronism.

Eric.