Carl's fluxgate Gradiometer ckt.

[Krzysztof]

Hi Damiafix.

If a damaged AD557 is out of the situation:
The outputs PIN6 - PIN13 SCL007 have full access to the results of the FGMs.
Simply take a small ATiny / PIC / STLite with access to the LCD and some code.
Already have a digital readout of nT / m * K.
The goldpins, which encouraged the buildings, allow the inclusion of such a snap to each MAG from Carl.
Best regards Chris.

[damiafix]

Dear Chris,
I'll make all the ceck and I'll let you know the results!

Many, many, many thanks, Dam.

[damiafix]

Hi Chris,
I'm sorry for the delay, I was busy with the work, and plus my little laboratory is into a hot garage car, night and day like a furnace!
I realized the "Logic Analyzer" that you have suggested me; I've tested it with 1 fgm-3, and it work perfectly, but I don't have still tried with the 2 sensors.When power on the circuit, after the 10 seconds of calibration, if I put close the sensor a metal, the led increment, start from the led connected to the SCL pin 13 (LSB) to the SCL pin 6 (MSB). The LSB (pin13) remain almost always on, sometime off, maybe it could to depend from the room where I make the tests, there was metals around.
I think that in the pdf of the FMX-1, written from Carl, there's something of wrong, between the schematic and the pratical parts placement on the PCB board.
In the schematic, the connections between the SCL007 and the AD557 are differents respect to the pcb parts placement. Please, see the follows pictures, altough I don't think that it's much rilevant, becouse the difference is on the number order of the SCL output wired to the DAC, but pratically it is right.

FMX-1 Parts Placement


FMX-1 Schematic


While, about the AD557 TEST, those are the results:

Test 2a:
Remove the SCL or make connections to AD alone.
Join your logic analyzer to the AD557.
All inputs PIN1 - pin8 connect to GND.
At PIN16 should be around 0V.

here:0.70v

Test 2b:
All inputs PIN1 - PIN7 connect to GND.
Pin8 connect the 5V.
On PIN16 should be around 1.25 V. (50% Vmax)

here:4.18v; if I remove the transitor Q1: 4.23v

I think that the DAC is damaged, what do you think?

Many thanks again!Dam.

[epitopios]

damiafix , you are right , you have eagle-eye
until now , I didnt notice !!!





regards , epitopios

[damiafix]

Hi Epitopios,
many thanks for your reply! Yes, it's so strange....your gradiometer, work correctly?

All the best, D.

[Krzysztof]

Hi Damiafix.
I think Carl looked at SCL006, SCL007 and wrote.
Test2b is bad. You can not cut corners. Apply your well-connected to the analyzer Ad557.
Measurement of the PIN14 & PIN15 & PIN16 connected together.
Test3a:
All inputs PIN1 - PIN8 connect to GND.
Test3b:
Inputs PIN2 - PIN8 connect to GND.
Pin1 connect the 5V.
Test3c:
Inputs PIN3 - PIN8 connect to GND.
PIN1 - PIN2 connect the 5V.
Test3d:
Inputs PIN4 - PIN8 connect to GND.
PIN1 - PIN3 connect the 5V.
Test3e:
Inputs PIN5 - PIN8 connect to GND.
PIN1 - PIN4 connect the 5V.
Test3f:
Inputs PIN6 - PIN8 connect to GND.
PIN1 - PIN5 connect the 5V.
Test3g:
Inputs PIN7 - PIN8 connect to GND.
PIN1 - PIN6 connect the 5V.
Test3h:
Inputs PIN8 connect to GND.
PIN1 - PIN7 connect the 5V.
Test3i:
PIN1 - PIN8 connect the 5V.
Test3j:
PIN1 connect the 5V.
Other Inputs connect to GND.
Test3k:
PIN2 connect the 5V.
Other Inputs connect to GND.
Test3l:
PIN3 connect the 5V.
Other Inputs connect to GND.
Test3m:
PIN4 connect the 5V.
Other Inputs connect to GND.
Test3n:
PIN5 connect the 5V.
Other Inputs connect to GND.
Test3o:
PIN6 connect the 5V.
Other Inputs connect to GND.
Test3p:
PIN7 connect the 5V.
Other Inputs connect to GND.
Test3r:
PIN8 connect the 5V.
Other Inputs connect to GND.
Good connection to the analyzer, ad557, GND and Vss ensure the stability of the test.
Best regards Chris.

[Krzysztof]

Hi Epitopios.
Whether your picture MAG Carl?
And how goes the MAG EPE?
Where the cables go from the center of images (SLC007-AD557)?
You measure something?
Best regards Chris.

[Krzysztof]

Hi Damiafix.

The LSB (pin13) remain almost always on, sometime off, maybe it could to depend from the room where I make the tests, there was metals around.

This is normal. It also means that MAG is sensitive!
FGM Any change in position on the table already in the area will change the value of the LSB.
In addition, the accuracy of digital output (each) is + / - 1LSB.
Your presence and position of the hand on the table (watch) have changed the result.
Recovers more precession in temperature and Micro changes intensity of Earth's field.
Expected accuracy is better than 1% at 8-bit measurement.
In the case of a program like Carl's MAG is good to do the calibration as often as possible. I always reset the background level and expand the measuring range at max!
I do not know but I guess the program algorithm. It is similar to the calibration of a compass. It measures the minimum and maximum field strengths and sets it on the range 0 - 255th.
Then the sensitivity is variable after each calibration!
Best regards Chris.

[damiafix]

Hi Chris,
I've removed the AD557 from the pcb and placed it on a breadboad for the test....



Those are the results for me:



Best regards, Dam.

[Krzysztof]

Hi Damiafix.
Almost perfect results.
As a laboratory of Analog Devices!
SCL and AD557 are OK!
Now search for defects in the soldering on the PCB.
The next test for the moment.
Best regards, Chris.

[Krzysztof]

Hi Damiafix.
We test the system V / f.
AD557 removed, the rest of the original PCB.
PIN14/15 to GND.
PIN16 let the tension forced another potentiometer with a value of less than 1Kom.
Cold end to GND.
Hot 1Kom end of the resistor to 5 volts.
Central to the PIN16.
Test4: Set the potentiometer R6 1Kom.
Force control voltage: 0V; 0,2V; 0,5V; 1,0V; 1,5V; 2,0V; 2,5V.
Test5: Set the potentiometer R6 2Kom.
Force control voltage: 0V; 0,2V; 0,5V; 1,0V; 1,5V; 2,0V; 2,5V.
Test6: Set the potentiometer R6 3Kom.
Force control voltage: 0V; 0,2V; 0,5V; 1,0V; 1,5V; 2,0V; 2,5V.
Results Listen!
In addition, you can see the results on the oscilloscope as in the description of Carl's.
Kind of a simple test, but control the voltage at PIN16!
You can choose the value of R6 for future work would always be the full range of audibility.
As you know, forcing PIN16 will always 0-2,5V.
Working conditions converter V / f Now you can choose!
Good luck. Chris

R6 to GND through resistor 100om - Security transistors.

[Krzysztof]

Hi Damiafix.
Out of curiosity.
Test7:
Measure the resistance between the AD557 PIN14 - PIN13.
Measure the resistance between the PCB blank PIN14 - PIN13.
Best regards Chris.

[damiafix]

Hi Chris,
many thanks, I'll ceck again the pcb!

All the best, Dam.

[damiafix]

Hi Chris,

H
Test7:
Measure the resistance between the AD557 PIN14 - PIN13.

1.62 KΩ

Measure the resistance between the PCB blank PIN14 - PIN13.

18 MΩ (from the dac socket without the IC)

About the test "systen V/f " I don't understood where I must to measure the results; maybe the frequency value on the VCO output (Pin 1 of the LM393)?
Unfortunately I don't have a frequency meter (I must to buy one, I know )for now, I could to use a frequency meter software? The frequency range, that I must to measure is very low (5-30hz) and from 0hz to 20.000 hz an pc audio card should to be suitable, right? For the output of the sensor is different, the pulse train is in the 100khz range...
What do you think?

All the best, Dam.

[Krzysztof]

Hi Damiafix.
Everything will be fine and free You do not need the frequency meter.
Your measurement is your ear, you hear the signal from the speaker.
Frequency easily done by a computer sound card. The simplest program "Winosc" I will when I find bearings in the net.
Then you will see the images from anywhere in the Part V / fi compare to figure Carl.
Test6 is there to find the best operating point V / f.
The resulting signal is the most useful and most famous in the 100Hz - 2kHz.
Above 4kHz means a very strong signal, or interference, hardly use it.
Tenderest place is at the lowest frequencies. However, assuming the presence of EMI at the beginning of 20Hz can be ignored as a disturbance, the rest is pure signal.
Of course, like Walk a little more, you can set a threshold of 0 V at 30Hz, and then you have the maximum sensitivity and auditory impressions like in a normal MD!
Best regards Chris.