Problem with EPE Magnetometry Logger

Maybe we'll get a response.

Crane

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,20982.msg148656.html#msg148656

Might be a thermal runaway voltage regulator or one that is out of spec. I used to see that with regulators feeding the old TTL chips sometimes.  Easy enough to check though.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,20982.msg148751.html#msg148751

GD

After a thorough re-check of all soldering, connections etc., I finally got the unit working well and very stable, before adding any resistor btw. pin 40 and gnd.
Probably, there was some hidden fault at some soldering point or a loose connection which was eventually fixed...

Should I still try adding a 33k (30k not available) or 22k, or 10k resistor, as a further enhancement of the stability of the pic function?

Can I try adding it and see what happens, without harming the pic's functions?
I got the advice not to add any resistor btw pin 40 and gnd, if I am not sure whether the pic's built-in pull-up resistors connecting the pin to +5V are activated. How could I test for it?

Looking forward to your reply.
lazzy

Posted on: March 02, 2011, 10:41:34 AM

---

On the other hand, I found out the following 'faults' and would like to share them with all of you out there who happen to have made the kit as well:

a) although I followed the exact circuitry plans and connection, soperation of leds D3/D4 seems to have somehow interchanged: D3 is activated with record switch S2, while D4 is flashing at the sensors sampling rate (once per sec); I tried to simply connect D3 to D4 pins and vice versa and they operate as planned, without any (visible) alteration in the pic operation

b) likewise, connecting the 'top' sensor to the small pcb shown at the top right of the plan in Fig.2 (and at the top of Fig.6), and the 'bottom' sensor to the pcb shown at the middle right in Fig.2 (and at the bottom of Fig.6), I got the readings from the 'top' sensor next to 'B' sign in the test display (appearing in the lower lcd line with test switch S4 'on'), and the readings of the 'bottom' sensor next to the 'T' sign; I 'cured' this by simply interchaning the X3/X4 sensor connections at the respective pins on the pcb; this way, I get correctly the readings of top sens. next to 'T' and readings of bottom sens. next to 'B', again without any (visible) alteration in the pic operation

for both the above 'faults' I have the same question: should I worry with possible problems due to the interchanges I made in connections? is there a way to test this?

moreover, could these 'faults' be an indication of a 'faulty' pic function because an incompatibilty of the original circuitry to the 'modern' pic877-20 I use? in that case, what should be changed in the circuitry in order to restore proper operation, with the relevant connections as originally planned?

apart from the above, apparently harmless, 'faults', the unit works well and quite stable;
one last thing I have to test is the connection to pc and data transfer... more on this and on the sensor alignment, soon...

lazzy

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,20982.msg150034.html#msg150034

GD,
Ok with the resistor issue.
What is your opinion about the 'faults' I've described?

Communication with pc/data transfer and memory clearance operation verified.
Would you think that using an usb-to-serial adaptor would work with pc comm?
I ask this because pc comm does show some instability (similarly with the resistivity logger), while serial ports are vanishing from pc's...

sensor alignment would be the next step.

and, talking about the sensors, Fig.5 and 6 of J.Becker article show both sensors to be positioned with their pins upwards; however, in the Application Notes for High Sensitivity Gradiometer supplied by Speake&Co, it is mentioned that "if you adopt ... hanging the gradiometer vertically, then it is an advantage to hang it from the end that makes the wires from each sensor emerge in the downward direction"; does this mean that, in a vertical probe, the sensors connection pins should be pointing downwards instead? 

rgds, lazzy

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,20982.msg150284.html#msg150284

joe,

for the pre-prog pic, look at magenta electronics in the uk:

You are not allowed to view links.
Please Register or Login

they also provide pics & full kits of other epe magazine constructional projects (data loggers etc.), sometimes with important operational enhancements of the original projects.

rgds, lazzy


Posted on: March 03, 2011, 03:26:56 AM

---

gd,

what exactly are the values measured from the sensors?

in p.471 (top of middle column) of the article is mentioned that "each pair of frequency count values is stored to non-volatile memory exactly as received";

on the other hand, p.469 (bottom middle col.) states that "the output is a ... rect. pulse whose period is directly proportional to the mag. field strength (giving a freq. which varies inversely with the field)";

finally, as stated in the Speake tech-specs, the sensor field range is +/-50 * 10-6 Tesla; the sensor actually measures flux density in gauss (1gauss =10-4 Tesla), while the typical frequency swing for the full range of the sensor is stated to be about 120000 to 50000 Hz;

the output values on the lcd range from 40000-50000 (in 'quiet' surroundings) to about 120000 (when approaching a small magnet to one of the sensors), so these values seem to express the frequency swing in Hz, as stated above?

in that case, would a displayed value of 60000 correspond to 1/60000 mag.field units, that is 1/60000 gauss, or 1/60000 *10-4 Tesla?

rgds, lazzy   

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,20982.msg150295.html#msg150295