Moodz' Awesome Gold Pulse Induction Version 3 - MAGPI V3 Project

[darov922]

Thanks Ian .. I just took it out for a test swing ... seems to be OK ... I am not sending the shafts .. just illustrating how I mount them.

2 x 20mm conduit clamps - as you can see no expense spared LOL. You will have to open the electronics case ( carefully ) to remount them.

moodz

This project is very interesting and I really want to make this detector... do you have any test video? I am looking for a detector to search for somewhat small gold nuggets in low mineralized soils. Will this detector work?

Dario

[moodz]

Thanks Moodz,
Appreciate all the help.

Thanks Dean .. dont get me wrong I am not trying to dismiss your ideas for improvement .. I just want to lock down the V3 hardware as there are some built up units already operating by various people and whereas we can change the software because that only involves re-programming those existing units ... if we change the hardware then those existing units and ones being built up might become obsolete and those people would not be happy.

For your info ... I have V4 in simulation already with NMOS and significantly higher power and V5 which is bipolar ... not sure if target sensistivity will increase that much though we hopefully will start to see a discrimination function.

moodz

[moodz]

This project is very interesting and I really want to make this detector... do you have any test video? I am looking for a detector to search for somewhat small gold nuggets in low mineralized soils. Will this detector work?

Dario

Hi Dario ... not big on videos however I will see if I can put something together. This detector will definitely work very well in low mineralized soils ... it will also work well in mineralized .. maybe not extreme ( needs software upgrade ).
It is good for small and big nuggets. There is no target hole ( as far as I know ) as some detectors have. The parts should be easy to come by. I have some boards available.

moodz

[darov922]

Hi Dario ... not big on videos however I will see if I can put something together. This detector will definitely work very well in low mineralized soils ... it will also work well in mineralized .. maybe not extreme ( needs software upgrade ).
It is good for small and big nuggets. There is no target hole ( as far as I know ) as some detectors have. The parts should be easy to come by. I have some boards available.

moodz

Thank you very much for your answer. I'll be waiting for those tables, I'll also wait a little longer until version 5, since I like a version with NMOS transistors better and if it's bipolar even better, I want to start buying the components anyway.

Dario

[Dean Sarelius]

Thanks Dean .. dont get me wrong I am not trying to dismiss your ideas for improvement .. I just want to lock down the V3 hardware as there are some built up units already operating by various people and whereas we can change the software because that only involves re-programming those existing units ... if we change the hardware then those existing units and ones being built up might become obsolete and those people would not be happy.

For your info ... I have V4 in simulation already with NMOS and significantly higher power and V5 which is bipolar ... not sure if target sensistivity will increase that much though we hopefully will start to see a discrimination function.

moodz

Hi Moodz,
Yes I uderstand and it is important that there is a baseline project that we work from otherwise it can get very confusing.
My intention it to complete the V3 and to make any changes that would eliminate the shortcomings of my previous designs hence the addition of the buffer.
In the oscilloscope image you provided previously I could see some distortion of the flyback, refer below image, which I have been able to eliminate partially by buffering the high gain stage.
A question regarding the X sample, to my understanding the X sample is taken during the magnetic field cycle and not the back EMF cycle or is my understanding wrong..?
Also I see a lot of advantages with bi-polar front end so was considering to design a detector around that front before you released the completed V3 schematic.
Do you think it will be better to wait for the release of the V4 or would you consider the V3 to be a solid design already..?
I have some hot rocks at the ready for testing the V3 so if it can ignore these then it will be happy days..

[moodz]

Hi Dean ... so all good ..if the buffer results in lower noise on the output then good. Most detector designs I have seen sample right off the output of the preamp with no buffering.
The X sample should occur as close as possible to the end of the flyback and for most coils the sweet spot is 1.1 to 1.8 microseconds.
With resistive sampling the width should be approx 0.1 microseconds but ppl can play with this to adjust for noise and sensitivity.

I found a lot of difficult targets with a GPZ7000 ... ( small nuggets, old buckshot, hotrocks ) ... The V3 is able to see them with about the same sensitivity.
The V3 is no better on hotrocks than the GPZ ... though the GPZ does not find alot of hotrocks.

moodz

[Dean Sarelius]

Hi Dean ... so all good ..if the buffer results in lower noise on the output then good. Most detector designs I have seen sample right off the output of the preamp with no buffering.
The X sample should occur as close as possible to the end of the flyback and for most coils the sweet spot is 1.1 to 1.8 microseconds.
With resistive sampling the width should be approx 0.1 microseconds but ppl can play with this to adjust for noise and sensitivity.

I found a lot of difficult targets with a GPZ7000 ... ( small nuggets, old buckshot, hotrocks ) ... The V3 is able to see them with about the same sensitivity.
The V3 is no better on hotrocks than the GPZ ... though the GPZ does not find alot of hotrocks.

moodz

Hi Moodz,
Thats awesome if your design is as good as the GPZ7000 then its happy days.
If you feel confident the V5 could be ready for release within the next few weeks then will prefer to hold off and lay out the board for a bi-polar design.
Have been running some simulations of the V3 and can see after approximatley 20-30ms the gate of the fet settles to around 0V so I gather this is the zero point you mention..?
cheers

[Dean Sarelius]

Hi Moodz,
After completing simulations with the original circuit there is only one major issue which has jumped out that needs to be considered.
Since this is a high side switch configuration the flyback has a tendency to reach 6 or 7 volts at the 1-2us region of the return pulse refer screen shot below of the pulse measured at the drain of BSP230 in my case this is the drain of the BSP304.
Referring to Channel A there are two markers Marker 1 (x1,y1) and Marker 2 (x2,y2).
Marker 1 (light blue) is the coil voltage just before switch off y1 = 4.7952V while Marker 2 (yellow) is the return voltage y2 = 6.4451V.
Just for reference the time difference between Marker 1 and Marker 2 is highlited with dx of 2.083us this was with a coil inductance 300uH, Rs 1.2ohm and Cp=100pf.
As you can see the return voltage at the 2us region is above the threshold of the supply rail to the op-amp which would necessitate increasing the the supply voltage to the LM6171, NE5534 and DG411 switches.
For reference have included screen shot pdf of the simulated circuit.
I am running simulations changing the supply voltage to 9V and will feedback my results shortly.



Moodz front end complete withou.pdf

[moodz]

Hi Dean I have scoped out the actual signals at the input / output to the preamp on the V3 for a complete transmit cycle ( 30us on / 30us off ) and the X sample pulse from the PIC32.

The top Light Blu trace is the X sample pulse.
The Pink trace is the input to the preamp.
The Darker Blu trace is the preamp output.

You should be able to see that the input to the preamp never exceeds 4.5 volts positive and 1 volt negative. This is because there are two diodes ( one damping and one protective to +5 volts ) across the input to the preamp.
So the input to the preamp never exceeds the supply rails and cannot due the diodes.
Hope this clarifies.

I can see you have gone to a fair bit of trouble to simulate the circuit. The coil I use is 0.5 ohms possibly 0.6 ohms. but this should not affect the result.

Cant tell from your results but was the damping control loop settled when you measured ?

moodz.

[Dean Sarelius]

Hi Dean I have scoped out the actual signals at the input / output to the preamp on the V3 for a complete transmit cycle ( 30us on / 30us off ) and the X sample pulse from the PIC32.

The top Light Blu trace is the X sample pulse.
The Pink trace is the input to the preamp.
The Darker Blu trace is the preamp output.

You should be able to see that the input to the preamp never exceeds 4.5 volts positive and 1 volt negative. This is because there are two diodes ( one damping and one protective to +5 volts ) across the input to the preamp.
So the input to the preamp never exceeds the supply rails and cannot due the diodes.
Hope this clarifies.

I can see you have gone to a fair bit of trouble to simulate the circuit. The coil I use is 0.5 ohms possibly 0.6 ohms. but this should not affect the result.

Cant tell from your results but was the damping control loop settled when you measured ?

moodz.

Hi Moodz,
I think I know where the issue is and it is with respect to charging the gate capacitor C1.
Without the bias there is no damping of the flyback hence the flyback is able to fall deeper than it should.
I can see on your scope the flyback is settling around zero volts where it should be however in my simulations it is passing the zero volt line and heading into the 6V region.
For some reason I am having issues to get the DG411 model to sample the pulse at 1.6us for the bias cap and am still working through it.
Hopefully will have it working soon.
Thanks again for your help..cheers

[Dean Sarelius]

Hi Moodz,
I have been unseccessfull in simulating your circuit within Multisim , just wondering if you have any LTSpice file which you can share perhaps..?

[Auto-Mation-Assist]

Hi Moodz,
I have been unseccessfull in simulating your circuit within Multisim , just wondering if you have any LTSpice file which you can share perhaps..?

I'm not sure if this will help but your simulation appears to be missing the isolation diode between the TX coil and the TX driver.

[Dean Sarelius]

I'm not sure if this will help but your simulation appears to be missing the isolation diode between the TX coil and the TX driver.

Yes I noticed that too and put the diode back in but without any change. The simulator prefers gear rather than trapazoidal so not sure if that could be causing some issues also..?
I have swapped out the DG411 with the ADG411 as well as with regular voltage controlled switches have tried long and short time intervals between interations always running extended siulations to capture the damping.
I think it could be an issue with Multisim so think I will try on another simulator maybe LTSpice, Proteous or Circuitmaker.
Will let you know if one of these are succesfull.

[Dean Sarelius]

Hi Dean I have scoped out the actual signals at the input / output to the preamp on the V3 for a complete transmit cycle ( 30us on / 30us off ) and the X sample pulse from the PIC32.

The top Light Blu trace is the X sample pulse.
The Pink trace is the input to the preamp.
The Darker Blu trace is the preamp output.

You should be able to see that the input to the preamp never exceeds 4.5 volts positive and 1 volt negative. This is because there are two diodes ( one damping and one protective to +5 volts ) across the input to the preamp.
So the input to the preamp never exceeds the supply rails and cannot due the diodes.
Hope this clarifies.

I can see you have gone to a fair bit of trouble to simulate the circuit. The coil I use is 0.5 ohms possibly 0.6 ohms. but this should not affect the result.

Cant tell from your results but was the damping control loop settled when you measured ?

moodz.

Hi Moodz,
Succesfully simulated your circuit with and without the buffer, I must say this is a very interesting circuit I would really like to know where you got the ZPD concept from..?
When comparing the results there doesnt appear to be much of a difference in the performance both circuits perform exactly as expected providing the much needed negative bias voltage for Q2.
The bias voltage settles at around -4V and it takes approximatley 90ms to reach this value with and without the buffer.
Below are screen shots results with the addition of a buffer using the NE5532 A and B one for the buffer the other for the TX feedback replacing the NE5534.
Since the buffered circuit works I will make these changes to my design and start laying out with the new schematic.
Simulations were completed with Multisim.
Screenshot 1
Blue = Bias Voltage Q2
Purple = TARG

Screenshot 2
Green = Output from NE5532AIP
Red = Input to LM6171AIM

[Dean Sarelius]

Hi Moodz,
Just laying out the schematic and note that the datasheet for the DG411 all the test circuits are applying the input to the source pins.
This would mean that the source pins P3(S1), P14(S2),P11(S3) and P6(S4) should be tied to the output of the opamp.
I will be making this change to my schematic also.
Suggest if you think all these changes deserve a new thread let me know and I will kick it off as MAGPI V3.1.
Cheers,
Dean