Atmospheric Electricity

[cecarnicom]

I am attempting to establish a quantitative measure of atmospheric electricity.
I have established the following algorithm to accomplish this and would like to
know if you can identify any flaws or failings in the reasoning process. I am new
to the topic of electrometers. Here are the major steps that I have outlined:

1. Construct an electrometer circuit based upon the use of a NJFET transistor.
The circuit will produce a full scale ammeter deflection of 50microamps at 9v.

2. Ground the electrometer by touching the gate lead of the NJFET.

3. Upon initial deflection of the ammeter, acquire time data (t) as a function
of even current readings (I) in microamps, e.g, 0, 10, 20, 30, 40, 50 microamps.

4. Assuming that the circuit can be accurately simulated in electronic simulation
software (e.g, Circuit Maker 6.0), determine gain levels at even current readings
(e.g, 10, 20, 30microamps etc) by using the ration Vd / Vg where Vd is the
voltage on the gate lead and Vg is the voltage on the drain lead.

5. Reduce the measure current values by the respective gain values at the even
current readings. Call these values (I')

6. Plot I' vs t.

7. Construct a polynomial regression of I'(t) of the form:
I' = at^2 + bt + c

8. Based upon the definition of I = dQ /dt

solve for the total charging current (Q') from:

Q' = integral [ I'(t) dt] with limits from t1 to t2.

9. Establish the charging rate per second as Q' / (t2 - t1)

10. Since 1 Amp = 1 Coulomb / sec, convert the charging rate in Coulombs to Amps.

Are there any problems or failures with this process that has been developed?

Thank you very much,

Clifford E Carnicom

 

[jbartos]

Suggestions:
1. There will have to be some sensor of the atmospheric electricity, perhaps a suitable antenna or lightning rod.
2. Some protection or current limitter of the input will be needed since the atmospheric electricity could become very strong at a time.
3. The charging current is normally referred to a capacitor/capacitance that still needs to be addressed in terms of voltage to withstand.

 

[cecarnicom]

jbartos,

Thank you for taking the time to reply. Actually I have already built a circuit and and in the process of testing - I am trying to ensure that my methods and results are correct. If it helps to send you the schematic, I will be glad to do that if you would like. In this circuit the gate lead is acting as the antenna or sensor. I have attached a 1 megaohm resistor on the end of it, which apparently is a protective measure for the jfet as you have described and suggested. A short wire has also been attached to act as an antenna.

My major concern is whether or not I am using the gain factor correctly and the subsequently developed algorithm corrrectly to compute the input current. The njfet gate lead is apparently regulated by voltage instead of current, and yet it is important to me to determine the input current since that is what I am trying to measure. My gain factors are based upon simulation software - I do not have any direct readouts from the circuit.

I have several test cases which are producing very consistent results, and the numbers appear to be reasonable ( e.g., approx. 12 microamps) but also a bit higher than is expected from my research.

I am very interested in any feedback on this from those who are knowledgeable in electrical engineering and theory. Regards, from Clifford.

 

[jbartos]

Suggestion/Comments marked by ///\\\:
I am attempting to establish a quantitative measure of atmospheric electricity.
///It appears that the quantitative measure is the current in time interval (t1,t2) since it is obtained via integration which results in the charge Q' (which is a number over interval (t1,t2), 1 second).\\ I have established the following algorithm to accomplish this and would like to know if you can identify any flaws or failings in the reasoning process. I am new to the topic of electrometers.
///Please, visit

http://www.eskimo.com/~billb/electrom/electrom.html

http://www.standardimaging.com/electrometers/premier.htm

http://www.fis.uc.pt/museu/120ing.HT

http://www.northcountryradio.com/elec.htm

http://www.standardimaging.com/electrometers/max4000.htm

etc. for information which I see behind your project, except the mathematical relationships.\\ Here are the major steps that I have outlined:
1. Construct an electrometer circuit based upon the use of a NJFET transistor.
///O.K., see one of the links above for similar application.\\The circuit will produce a full scale ammeter deflection of 50microamps at 9v.
///O.K. It could potentially also display electrical charge. in C/sec.\\2. Ground the electrometer by touching the gate lead of the NJFET.
///O.K., see one of the above links, for the similar approach.\\3. Upon initial deflection of the ammeter, acquire time data (t) as a function
of even current readings (I) in microamps, e.g, 0, 10, 20, 30, 40, 50 microamps.
///O.K., and potentially in nC/sec.\\4. Assuming that the circuit can be accurately simulated in electronic simulation
software (e.g, Circuit Maker 6.0), determine gain levels at even current readings
(e.g, 10, 20, 30microamps etc) by using the ration
///Please, clarify.\\ Vd / Vg where Vd is the voltage on the gate lead and Vg is the voltage on the drain lead.
///Clarify Vd versus drain lead and Vg versus gate lead.\\5. Reduce the measure current values by the respective gain values at the even current readings. Call these values (I')
///O.K.\\6. Plot I' vs t.
///O.K.\\7. Construct a polynomial regression of I'(t) of the form:
I' = at^2 + bt + c
///O.K.\\8. Based upon the definition of I = dQ /dt
solve for the total charging current (Q') from:
///Clarify the total charging current (Q') in terms of the Charge Q' in C/sec rather than Current in microAmps.\\Q' = integral [ I'(t) dt] with limits from t1 to t2.
///O.K.\\9. Establish the charging rate per second as Q' / (t2 - t1)
///O.K.\\10. Since 1 Amp = 1 Coulomb / sec, convert the charging rate in Coulombs to Amps.
///Clarify Coulombs to Amps in terms of C/sec to Amps.\\Are there any problems or failures with this process that has been developed?
///Pending the outcome of clarifications.
Any material may be forwarded to:
P.O. Box 2055, New York, NY 10159-2055, USA or faxed to:
718-721-5298 or 212-366-3999.\\\