EMF Probe Project

Thread Starter

Butterworth

Joined May 6, 2009
135
This appears to be it:
http://www.zen22142.zen.co.uk/Circuits/Testgear/emfprobe2.htm

EDIT: I think his simulation is flawed.
Thats the schematic it was based on all right. If the probe needs to offset 147v over the v/2, can I adjust the 470k resistor to match the voltages I need so it will not saturate at the rail?

EDIT: I found, what I believe to be, the manufacturer's test results on the input offset voltages per model of the TL084, but it says it was based on Vcc+ @ 15v(+/-) I have attached the pdf page of their testing. If this is true, then they claim the offset voltage on the input is about 3mv typ and 6mv MAX.
 

Attachments

Last edited:

Ron H

Joined Apr 14, 2005
7,063
Thats the schematic it was based on all right. If the probe needs to offset 147v over the v/2, can I adjust the 470k resistor to match the voltages I need so it will not saturate at the rail?

EDIT: I found, what I believe to be, the manufacturer's test results on the input offset voltages per model of the TL084, but it says it was based on Vcc+ @ 15v(+/-) I have attached the pdf page of their testing. If this is true, then they claim the offset voltage on the input is about 3mv typ and 6mv MAX.
Those aren't really test results. That's the datasheet, and it's where I got my input offset voltage information.
If you want to reduce the 470k far enough to prevent saturation, your gain will suffer severely. You can't make up the gain in the second stage without losing bandwidth.
I think I know another way, where you could avoid the 100uF cap, but it would still require a re-layout of your board. If you want to see that idea, let me know. It's still half-baked, so it might not turn out.:rolleyes:
 

Thread Starter

Butterworth

Joined May 6, 2009
135
Those aren't really test results. That's the datasheet, and it's where I got my input offset voltage information.
If you want to reduce the 470k far enough to prevent saturation, your gain will suffer severely. You can't make up the gain in the second stage without losing bandwidth.
I think I know another way, where you could avoid the 100uF cap, but it would still require a re-layout of your board. If you want to see that idea, let me know. It's still half-baked, so it might not turn out.:rolleyes:
To be honest, my board is already garbage due to the mistakes I carelessly made. I have no issues making a new board (I am designing a new layout as we speak) so any ideas are welcome and appreciated! :)
 

Ron H

Joined Apr 14, 2005
7,063
OK, I simulated this, and it worked. I added 6mV of offset to the input amp, and it still worked as expected. I changed some part values, so consider this to be a new schematic.

EDIT: Added 10pF cap across R1 to compensate for probe self-resonance.
 

Attachments

Last edited:

Thread Starter

Butterworth

Joined May 6, 2009
135
OK, I simulated this, and it worked. I added 6mV of offset to the input amp, and it still worked as expected. I changed some part values, so consider this to be a new schematic.
Interesting, so this worked on the simulation? What results could I expect in frequency ranges?

Thank you very much for taking the time to correct this schem for me! I will have to frequent the supply store tomorrow during my lunch hour for the new parts. I noticed you took out R7 from the v/2 line, as well as re-arranged op amp #3 to piggy back off of op amp 1, was this to boost the signal from the probe?
 
Last edited:

Ron H

Joined Apr 14, 2005
7,063
RF is a very poorly defined term. 1MHz is definitely RF. For example,the AM radio broadcast band is from approximately 535kHz to 1.6MHz.
The circuit should have about the same capabilities as your original circuit would have if it did not have the potential offset voltage problem.
 

Ron H

Joined Apr 14, 2005
7,063
Interesting, so this worked on the simulation? What results could I expect in frequency ranges?

Thank you very much for taking the time to correct this schem for me! I will have to frequent the supply store tomorrow during my lunch hour for the new parts. I noticed you took out R7 from the v/2 line, as well as re-arranged op amp #3 to piggy back off of op amp 1, was this to boost the signal from the probe?
The op amp I moved was originally configured as a voltage follower. It was redundant. I used it instead to compensate for the offset voltage of the input op amp. The gain from input to output remains the same.
 

Thread Starter

Butterworth

Joined May 6, 2009
135
The op amp I moved was originally configured as a voltage follower. It was redundant. I used it instead to compensate for the offset voltage of the input op amp. The gain from input to output remains the same.
Ok, I will pick up the required items to make this new circuit on my bread board and see how it goes tonight!

I will post my (real world) results when I have layed it out.

Stay tuned!
 

Thread Starter

Butterworth

Joined May 6, 2009
135
The high frequency, high gain circuit might oscillate if you make it on a breadboard.
What can I do to make this circuit without wasting time soldering, if not the bread board?

EDIT: For the nF cap I can assume ceramic type is acceptable?
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,248
What can I do to make this circuit without wasting time soldering, if not the bread board?
A breadboard circuit usually has a tangled mess of long wires with high capacitance between them. Frequently the contacts on a breadboard are intermittent.
The rows of contacts have high capacitance between them.
The wires and contacts have plenty of series inductance.

A compact stripboard or pcb construction should be used.
Each solder joint takes 1 second or 2 seconds to do.

For the nF cap I can assume ceramic type is acceptable?
A 0.1uF ceramic capacitor is good for the 100nF supply bypass capacitor.
 

Ron H

Joined Apr 14, 2005
7,063
Do you think a small cap across R1 (while he's proto boarding) will help?
I had a 10pF cap across R1 in a previous version, but I guess it fell off.;)
There should definitely be a 10pF cap across R1. I will edit the schematic.
Thanks, CDRIVE!
 
Top