NEW !!! Full differential PI front end.

[Unregistered]

Why are you so aggressive!
I just wanted to learn moodz intentions about the future of his work. I am not the one promising to publish a field trial ready circuit.
BTW thanks a lot Moodz and Bugwiskers. You shared a lot in this forum.

If you have something as good as he has and you have spent a a lot of time and money why shouldn't he go commercial and get something back for his efforts?
I think he has shared a lot already don't you?
And there are those that don't want it to go commercial for obvious reasons!!!
I think that he and moodz should get together and each use the advantages of each others technology.

[moodz]

http://www.hayselectronics.com/parts.htm

Vasster .. your the man.

thanks very much .. moods

[moodz]

updated circuit ....

v303a

enjoy ... moodz.



NO TARGET

20 cent coin

[moodz]

There have been some changes as previously mentioned ....

Isolating diodes have been added. There is a gotcha with differential circuits and that is when noise gets into one side of the circuit and the will not reject it as it is not common mode. The 1n4004 diode on the mosfet will reverse bias as the flyback pulse falls ... why because ... the MOSFET looks like a capacitance to ground when it is off and this will charge via the diode ... when the flyback falls lower thant the peak charge the diode will reverse and the junction capacitance will isolate the relatively huge transfer capacitance of the MOSFET.

The CPU I am using ( dsPIC ) can easily drive the MOSFET directly ... however there is a relatively huge transfer capacitance from the gate to the drain and this will couple RF hash on the gate lead to the driven side of the differential circuit again unbalancing it and causing differential noise. NOTE ... this would also impact standard PI designs ... so check for noise entering via this path. The LM311 is a quick and dirty way to achieve high voltage ... high current drive and no RF hash from the CPU drive line.

Likewise D6 provides isolation of the coil after the Tx pulse ends. It also provides a bias point for the clamping diode system on C4

Notice that the opamp is fully isolated from DC at the inputs. After all we are amplifying pulses ( AC ) here. The opamp will bias nicely to approx mid v+ and you can see this in the scope pics in the previous post.
When the circuit is first turned on it may take 10 seconds or so to stabalise as C5 has to charge via R6 to provide the bias point.

At the moment the only voltage regulator is the +5 volt reg for the microprocessor. ( not shown in the schematic )

The coil I am using is fairly ordinary so I will be looking to speed this up next .... however I feel that the output is already good enough to feed into the A2D converter in the CPU for processing.

will post more info ....

any suggestions welcome.

moodz

[Carl-NC]

Notice that the opamp is fully isolated from DC at the inputs.

You still need a DC path on the + input.

[moodz]

You still need a DC path on the + input.

Hi Carl .... its buried away in the data sheet for the CA3130 ( page 7 on the intersil specs ) ... to quote ...


DIFFERENTIAL DC VOLTAGE
(ACROSS TERMINALS 2 AND 3) = 0V
OUTPUT VOLTAGE = V+ / 2

If I isolate all inputs with low loss caps then this works.
So i dont need any biasing to ground. The DC value across the inputs is 0.
So this amp has no DC gain and the caps appear as very high impedances DC but low impedances to the pulse waveforms.

I have used this trick in a number of places.

Another example of this type of self biasing is to take a hcmos inverter and put a 1M feed back resistor from the output to the input. The output will settle to VCC / 2 and you can use the device as a type of crude amplifier.

Regards,

moodz.

[simonbaker]

Hi Moodz:

What is the value of the damping resistor? It looks like 470R....?

Regards,

-SB

[Unregistered]

It reads 470R to me

[moodz]

Hi Moodz:

What is the value of the damping resistor? It looks like 470R....?

Regards,

-SB

Simon ... 470 ohms.

I have also tried placing the damping resistor right at the coil .... seems to have a marginal benefit.

moodz.

[moodz]

descrimination for moodzPI

... well the adventure continues ...

The last frontend v303b was perfectly good as a frontend for a standard PI however the ability to analyse the conductivity of the target would be very handy ..... some might call it descrimination.

So ... v400a is posted below. Note that the 'clamping diodes' have been replaced by two 5.6 volt zeners. The amplifier now has a gain of approx one. ( yes 1 ) and for those who build this circuit and examine the waveform out of the amp you will see very interesting behaviour at the peak for ferrous vs more conductive targets.
The post processing will now take place at two points ... at the peak and the base. ( ie two main channels to the cpu )
For those who cant be bothered building the circuit yet well ... Ok ... Ill post some scope pics on the next post. ( gotta get my points up for Carl's comp after all. )
cheerz all.

[moodz]

descrim pics for moodzPI v400a

well here they are ...

First one below is NO TARGET. o/p from amp in v400a schematic.


Second below is copper target. Note the base response but the peak is a little lower ( not much change )


Third is below with steel target. Note BOTH the base and the peak get 'fatter' ... now you should be able to guess where the two places where the post processor will be sampling.


cheerz from moodz.

[moodz]

Oh yeh ...
Both targets 10 inch square.
Tx pulse = 100uS
Coil the same one as per schem.

moodz

[moodz]

ooh ... it was already invented by Aziz

Aziz ... I had not seen this thread till now .... you already proposed the idea of differential front end.

http://goldprospecting.invisionplus....showtopic=2227


However the simple diode clamping proposed works ok in SPICE but not in practice as the +ve and -ve flybacks are not exactly balanced.

Good work ... I think that this should be pursued anyway and I agree with all your analysis.

Regards,

Moodz.

[Aziz]

Hi moodz,

Aziz ... I had not seen this thread till now .... you already proposed the idea of differential front end.

http://goldprospecting.invisionplus....showtopic=2227


However the simple diode clamping proposed works ok in SPICE but not in practice as the +ve and -ve flybacks are not exactly balanced.

Good work ... I think that this should be pursued anyway and I agree with all your analysis.

Regards,

Moodz.

Please keep on good working. You have the instruments for this and you are quick in doing this... And you made more improvements on the idea.

I am looking forward to your results..
Regards,
Aziz

[moodz]

Thanks ...

Hi moodz,



Please keep on good working. You have the instruments for this and you are quick in doing this... And you made more improvements on the idea.

I am looking forward to your results..
Regards,
Aziz

Thanks Aziz I honestly had not seen the circuit you published and I have come up with almost the identical design ... unbelievable !! I read somewhere that if you come up with an original design then someone somewhere else in the world will be working on similiar thoughts at the same time ... this happens in science many times ...I dont know if anyone has studied it ... it would make a good thesis. Maybe there is a collective consciousness ..
I will continue to work on this but the credit for any innovation must go to you .

regards,

paul.