Countering EMI in Pi detectors?

[Doug]

Countering EMI in Pi detectors
I have started  this thread (as suggested by  mick) so that we can discuss this issue without getting it mixed up the Micro control for SD 2000/2100? thread!
doug

[mikebg]

Countering EMI in Pi detectors
I have started  this thread (as suggested by  mick) so that we can discuss this issue without getting it mixed up the Micro control for SD 2000/2100? thread!
doug

If the participants of our forum think that term "PI detector" means PI device using monocoil, there is nothing for discussion. The monocoil has maximal sensitivity to interferences and generates maximal AIR & GND signal.

[Doug]

Countering EMI in Pi detectors
I have started  this thread (as suggested by  mick) so that we can discuss this issue without getting it mixed up the Micro control for SD 2000/2100? thread!
doug

If the participants of our forum think that term "PI detector" means PI device using monocoil, there is nothing for discussion. The monocoil has maximal sensitivity to interferences and generates maximal AIR & GND signal.

An anti interference coil is not mono coil. Its a mono TX with a fig 8  Rx phase / anti phase.With later ML's you can also run a DD in cancel mode.
doug

[Mechanic]

Countering EMI in Pi detectors
I have started  this thread (as suggested by  mick) so that we can discuss this issue without getting it mixed up the Micro control for SD 2000/2100? thread!
doug

If the participants of our forum think that term "PI detector" means PI device using monocoil, there is nothing for discussion. The monocoil has maximal sensitivity to interferences and generates maximal AIR & GND signal.

Hi Mike,

Pi detector means pi device using coil that goes as deep as or deeper than monocoil.....

And yes I believe that some sort of anti interference configuration is necessary, but something better than the current style as to not loose any depth.

Cheers Mick

[Doug]

Countering EMI in Pi detectors
I have started  this thread (as suggested by  mick) so that we can discuss this issue without getting it mixed up the Micro control for SD 2000/2100? thread!
doug

If the participants of our forum think that term "PI detector" means PI device using monocoil, there is nothing for discussion. The monocoil has maximal sensitivity to interferences and generates maximal AIR & GND signal.

Hi Mike,

Pi detector means pi device using coil that goes as deep as or deeper than monocoil.....

And yes I believe that some sort of anti interference configuration is necessary, but something better than the current style as to not loose any depth.

Cheers Mick

What about an asymmetric fig 8 coil ie  a big loop and a small loop with the output of the small loop amplified to the same level as the big loop?
doug

[Mechanic]

Hi Doug,

I was thinking a second coil, receive only, either the same size or smaller fixed about a foot above the tx/rx coil. The signal from this coil would then be fed to an amplifier inverted and added to the tx/rx signal to cancel the emi. Much the same as cancel mode on the ml's

Target signal will stay intact because of the space between the 2 coils, the target signal will never be the same strength at the top coil compared to the bottom coil.

Also with the coils being fixed the emi signals will always be the same in both coils even when the coil assembly is tilted, just a shame it would be a bit cumbersome!

Cheers Mick

[Doug]

Hi Doug,

I was thinking a second coil, receive only, either the same size or smaller fixed about a foot above the tx/rx coil. The signal from this coil would then be fed to an amplifier inverted and added to the tx/rx signal to cancel the emi. Much the same as cancel mode on the ml's

Target signal will stay intact because of the space between the 2 coils, the target signal will never be the same strength at the top coil compared to the bottom coil.

Also with the coils being fixed the emi signals will always be the same in both coils even when the coil assembly is tilted, just a shame it would be a bit cumbersome!

Cheers Mick

You could to some extent overcome the depth disadvantages of the fig 8 anti phase coil if  you have a gap between the coils ie 2 mono's spaced apart and TX on both coils and RX anti phase? But the coils would be very big!!
doug

[mickstv]

Hi All,  I've been messing with the Diff coil design that Moodz came up with,  not his latest design but the older version without the descrim setup.  It seems to do a very good job of canceling out interference like 50hz.  The only downside so far has been the speed of the diff coils I'm making,  the fastest I've been able to make still takes 10-12us before sampling.  Hopefully I can cure that problem with some rolls of litz wire that I've ordered from ebay.



Mick 

[Doug]

Hi All,  I've been messing with the Diff coil design that Moodz came up with,  not his latest design but the older version without the descrim setup.  It seems to do a very good job of canceling out interference like 50hz.  The only downside so far has been the speed of the diff coils I'm making,  the fastest I've been able to make still takes 10-12us before sampling.  Hopefully I can cure that problem with some rolls of litz wire that I've ordered from ebay.



Mick 

What pulse length, peak coil current and coil inductance are you using to get the 10-12u secs sample delay?
doug

[mickstv]

Hi All,  I've been messing with the Diff coil design that Moodz came up with,  not his latest design but the older version without the descrim setup.  It seems to do a very good job of canceling out interference like 50hz.  The only downside so far has been the speed of the diff coils I'm making,  the fastest I've been able to make still takes 10-12us before sampling.  Hopefully I can cure that problem with some rolls of litz wire that I've ordered from ebay.



Mick 

What pulse length, peak coil current and coil inductance are you using to get the 10-12u secs sample delay?
doug

Hi Doug,  the pulse length tested was 50us and also 100us.  I haven't actually checked peak current and coil inductance I just slapped it all together as an experiment.  From memory the 50us was developing about 280v p-p but I'll have to check that tomorrow.  The current coil I'm using is a spider wound coil using one pair of wire out of a length of cat5 cable the inner dia is about 18cm and 26 turns.



Mick

[WM6]

Hi Doug,

I was thinking a second coil, receive only, either the same size or smaller fixed about a foot above the tx/rx coil. The signal from this coil would then be fed to an amplifier inverted and added to the tx/rx signal to cancel the emi. Much the same as cancel mode on the ml's

Target signal will stay intact because of the space between the 2 coils, the target signal will never be the same strength at the top coil compared to the bottom coil.

Also with the coils being fixed the emi signals will always be the same in both coils even when the coil assembly is tilted, just a shame it would be a bit cumbersome!

Cheers Mick

Hi Mick,

try this design:

[Mechanic]

Hi Wm6,

With the 2nd rx coil mounted vertical, the emi signal it receives will be different compared to the 1st rx coil, some signal will be nulled in the reception due to its vertical mount, also it will be receiving emi signals that the primary won't be, so instead of the signals being nulled they will be added....

Hi Mick,

I have been watching Moodz' differential coil project closely. One thing that still bugs me is : emi will be the same but opposite in both coils, which will cancel the emi superbly, but in my mind the target signal should also be the same but opposite also, so why is it not cancelled?

One experiment I did with my 2000 was set up a timing that had a 200us pulse, and then 1 sample of say + polarity, then another 200us pulse, exactly the same sample delay and sample width, but of opposite polarity, with the earthfield trimmer set to null samples of exactly the same length. This should have cancelled the response from any target, but to my surprise I could still detect a 20c coin at about 30cm!! This was due to a slight imbalance somewhere in the cct, I think the earth field trimmer may have been set ever so slightly wrong. So what I'm getting at is if a slight imbalance exists, not everything will be cancelled. Perhaps this would sort of explain why the diff coil seems to work? Don't know, I will have to build one and test it to find out. Oh God! Another project Whats wrong with me!

Cheers Mick

[mickstv]

Hi Mick,  

When I get a chance today i'll have a look at the inputs to the diff amp and see how they compare on the cro.  The circuit I have build so far is only the diff coil and amplifier then into one sample channel and the only amplification is the THAT1510 which is running at about 400x.  When I was monitoring the output of the sample channel I was seeing 50c coin at about 25cm so with some more amplification post sample I should be able to get a better response.  Also tested various ferrite cores and these showed a negative response on the cro.



Mick

[Aziz]

Mick,

the diff coil doesn't cancel the EMI. It only cancels the EMI on the coil leads up to the differential input.

Any slight inbalance will reduce the CMRR. More on high frequency range.

Cheers,
Aziz

[Mechanic]

Mick,

the diff coil doesn't cancel the EMI. It only cancels the EMI on the coil leads up to the differential input.

Any slight inbalance will reduce the CMRR. More on high frequency range.

Cheers,
Aziz

Hi Aziz,

Why does it not cancel the emi? An AI coil does and it is a figure 8 style receive coil. Only difference is the cancellation is done in the coil itself, whereas the diff coil presents a differential signal to the diff opamp, should not the emi signal be the same but opposite?

The emi problem does not seem to be caused by the coil lead, if so the problem would not become worse with bigger coils and Ai coils would not work.....

Cheers Mick

[Aziz]

Mick,

the figure-8 coil magnetically induces different polarity signals (different winding orientation), which are added together (in series connected coil parts).
Vemi   = V1emi + V2emi,
V1emi =-V2emi (emi induction)
Vemi   = -V2emi + V2emi = 0  (perfect)

Hence subtracting them to 0.

Targets nearby the figure-8 coil produce different strong response (depending on the position) and unfortunately a substantial part of it will be subtracted by the other coil part.

A differential coil however doesn't cancel the magnetically induced emi, which unfortunatelly will be seen as a differential signal.

But there are other aspects as well: the electro-magnetic coupled emi (em wave), when the coil and it's leads can be seen as an antenna.
I didn't think about this further. Who knows?

Cheers,
Aziz

[Aziz]

To understand the whole RFI/EMI shielding I need to read a lot I think (I do not know and understand everything guys!).
What makes it difficult to imagine is as follows:
A piece of conducting wire (even a straight line wire) is an inductor and capacitor!

Even the piece of wire doesn't look like a coil, it can induce magnetic and electric field coupled signals.

Only a super conducting shielding material is capable to block all RF/EM interference.
Aziz

[IBGold]

Hi Doug,

I was thinking a second coil, receive only, either the same size or smaller fixed about a foot above the tx/rx coil. The signal from this coil would then be fed to an amplifier inverted and added to the tx/rx signal to cancel the emi. Much the same as cancel mode on the ml's

Target signal will stay intact because of the space between the 2 coils, the target signal will never be the same strength at the top coil compared to the bottom coil.

Also with the coils being fixed the emi signals will always be the same in both coils even when the coil assembly is tilted, just a shame it would be a bit cumbersome!

Cheers Mick

Hi Mick,

You mean something like this Aziz's Coaxial coil.

Regards, Ian.

[Doug]

Hi Doug,

I was thinking a second coil, receive only, either the same size or smaller fixed about a foot above the tx/rx coil. The signal from this coil would then be fed to an amplifier inverted and added to the tx/rx signal to cancel the emi. Much the same as cancel mode on the ml's

Target signal will stay intact because of the space between the 2 coils, the target signal will never be the same strength at the top coil compared to the bottom coil.

Also with the coils being fixed the emi signals will always be the same in both coils even when the coil assembly is tilted, just a shame it would be a bit cumbersome!

Cheers Mick

Hi Mick,

You mean something like this Aziz's Coaxial coil.

Regards, Ian.

  looks good!
doug

[WM6]

Nice pro work Ian.

What is the "light"  of this coil?