Project: Oscilloscope AC Current Probe

Thread Starter


Joined Jan 18, 2008
This project began as a diversion during development of a power supply to control the charge voltage of a large bank of capacitors (~1.0 F) for a capacitor-discharge welder. The initial design of the power supply simply used a load resistor, relay, and comparator to attain that goal, and it worked with the relay pulsing at about 0.5 Hz. A second design in which the relay was replaced with a mosfet and higher frequencies were used to provide tighter voltage control did not work until I added an inductor, as in conventional switching power supplies. Out of pure curiosity, I wanted to observe and study current patterns associated with that inductor. This project remains one primarily for the bored and curious. It is not intended to be particularly quantitative nor something you will use everyday.

My design follows closely the design by Dick Cappels. He provides an excellent discussion of the theory and calculations needed for a current probe. His design uses a simple current transformer made from a toroid through which the current carrying conductor passes. My design goals were:

1) A split probe, so the current carrying conductor would not need to be interrupted.
2) A small probe that could be used in our smaller circuits.
3) A frequency response above 1 KHz that would be useful for analysis of PWM motor control circuits and switching regulators. Neither the low frequency response achieved by Cappels with a large inductance probe nor a DC response using a combination of current transformer and Hall device, as described in various patents, was needed.
4) Use parts that were readily available and characterized.

Probe Construction


1) Ferrite tubular bead (DigiKey 240-2296-ND; Steward 28B0375-400; 9.53 mm OD X 5.08 mm ID X 4.83 mm long)
2) Flat piece of scrap ferrite from a transformer E-core
3) 32 AWG magnet wire (about 1 m)
4) Hayes small hobby clamp (Hayes 104; Tower Hobbies LXK854)
5) Assorted wet/dry abrasive finishing paper (grits of 400 and 1200 or finer)
6) Assorted heat-shrink tubing
7) Wire for leads


1) The first step is to remove a small segment from the ferrite bead to give a C-shape. Working with ferrite is a little like working with glass or ceramic. Some grinding can be done dry, but ferrite tends to crack and chip easily. A soft, coarse grinding wheel running dry on a bench grinder can be used to remove the segment. Alternatively, a high-speed hand tool (e.g., Dremel, Racine, WI) with a carborundum cut-off wheel works well. Use the non-reinforced wheel (Dremel # 409) and keep it wet. If it runs dry, it will glaze the cutting edge, which creates more heat and cracks the ferrite. I put a small, water-soaked sponge on the exhaust side of the cut-off wheel and cut through both the sponge and ferrite simultaneously. The sponge keeps the wheel plenty wet, and its position helps control the mess produced by grinding the ferrite.
2) Remove as little of the ring as practical. The final shape should be a segment with more than 180° of the loop (Figure 1).
3) Cut a small piece of flat ferrite to serve as a “keeper” across the open throat of the C-ring.
4) Polish the faces of the C-ring and keeper to get as close a fit as possible. The finest grits of finishing paper can be obtained from automotive paint stores. Grits finer than 600 are preferred for the final polish. The probe describe here was finished using 1200 grit abrasive. Use plate glass as backing for the abrasive paper and water as lubricant.
5) Wrap 40 turns of 32 AWG magnet wire around the C-ring and finish in your favorite method. I used a pliable liquid cement (Pliobond; W.J. Ruscoe Co., Akron, OH) with shrink wrap for durability.
6) Add leads to the coil for connection to the amplifier.
7) Prepare the Hayes hobby clamp as shown in Figure 1. The clamp is a glass-reinforced plastic (not carbon fiber) and cuts easily with woodworking tools. A small recess was cut in each arm of the clamp to reduce the closed dimension and to help position the semicircular ferrite segment. A straight router bit (3/8” diameter) or end mill works well for cutting the curved recess.
8) Assemble the completed components as shown in Figure 1. The Hayes clamp maintains good alignment of the C-ring and keeper. Nevertheless, the keeper was mounted on a small piece of soft, double-sided tape to allow a little movement and to optimize contact between the faces. The C-ring part was mounted using cyanoacrylic adhesive (“CA”) and catalyst.


The A\(_{L}\) of the ferrite bead was determined by constructing a test coil of 20 turns on an uncut ferrite core, which gave an inductance of 200 μH (See: Cappels).


L = A\(_{L}\) N\(^{2}\)


A\(_{L}\) = 0.5 μH/turn squared

A 40-turn coil on the same form should have an inductance of 800 μH. The initial open-C inductance was about 28 μH without the keeper. With a coarsely finished keeper, that value rose to about 250 μH, and with polished faces on the C-ring and keeper, the inductance was 700 μH (about 88% of theory for a closed ring).

The next step was to calculate the low-frequency corner frequency (Fc) of the probe. That is the frequency at which there is a 3 dB (30%) loss in sensitivity (See: Cappels).

F\(_{c}\) = R\(_{total}\)/2πL

R\(_{total}\) = R\(_{coil}\) + R\(_{term}\)

In this case, the coil resistance was 0.32 Ω and termination resistance (R\(_{term}\); See:Schematic) was 1.1 Ω for a total of 1.42 Ω. The calculated F\(_{c}\) is: 323 Hz.

Probe sensitivity (i.e., output volts per ampere) is a function of number of turns and termination resistance. One can increase sensitivity by decreasing the number of turns or by increasing the termination resistance. Decreasing turns increases sensitivity in direct proportion, but decreases inductance and worsens low-frequency response as the inverse square. Increasing termination resistance improves sensitivity, but reduces low-frequency response in direct proportion. In the design presented here, those trade-offs were made based on the need for a small probe and an F\(_{c}\) no higher than 300 Hz. The probe can be used directly with an oscilloscope as described by Cappels, but addition of an amplifier increases sensitivity with little or no loss in low-frequency response.

Amplifier Design


C1 2.2 pF/50V (805)
C2 10/35V (1206) (DigiKey 587-1352-1-ND)
C3 10/35V (1206) (DigiKey 587-1352-1-ND)
C4 0.01/50V (805)
C5 2.2 tantalum/35V (LOESR, DigiKey 478-3105-1-ND)
C6 0.01/50V (805)
C7 2.2 tantalum/35V (LOESR, DigiKey 478-3105-1-ND)
IC1 ICL7660
IC2 LM6171
R1 1K
R2 51R
R3 22R
R_BIAS 100K (adjust as needed)
CT1 current transformer (See: Probe)
PCB See: Attached BRD file

Note: All capacitors are ceramic, unless noted; all capacitor values are μF, unless noted; all components are SMD; resistors are 805 packages.


The amplifier was constructed using common PCB techniques. EAGLE files for schematic and board are in the attached zip file. The completed board is shown in Figure 2 along with a non-editable version of the schematic. The little bug on top of the LM6171 is the 100K bias resistor. When the PCB design was etched, design of the amplifier was not finalized. Larger input resistors (R3 and Rterm) were used, gain was lower, and there was almost no offset. Changes made in the final design necessitated the bias resistor. The files attached here incorporate those changes. Also, in construction of the board, I did not have 805 packages for all components and squeezed 1206 components in their places. The board design and pad spacing, however, are for 805 components. The final PCB was shrink wrapped for protection. The three-pin header is used as the on/off switch, which is controlled with a jumper.

Performance Validation

Figure 3 shows the frequency response of the probe and amplifier to a sine wave current over the range of 300 Hz to 307 KHz. Note that Fc (corner frequency) is close to the predicted value of 323 Hz. Figure 4 shows qualitative performance for triangle and square waves over the same frequency range. The top tracing of each screen shot is amplifier output; the bottom tracing is the input waveform. Current was approximately 100 mA for all tracings (5.1V across a 51 ohm resistor). Usable tracings were observed with 20 dB (10 mA) attenuation on the signal generator. Within the range of most interest to me – 1.2 KHz to 200 KHz – output is a good approximation of the current waveform being monitored.

Potential Design Changes

1) Different op-amp? I am open to any suggestions for other op-amps or instrumentation amplifiers. Selection of the LM6171 was based on its high slew rate (3600 V/μS), unity-gain-bandwidth product (100 MHz), and ±15 V power supply.
2) Use a larger PCB with changes to facilitate trimming both offset and amplification factor.
3) Delete negative voltage supply and just use 2 batteries. Supply noise is obvious in the oscilloscope tracings.
4) Design changes to reduce total probe current. Current design runs at about 4.85 mA.

Have fun.

Last edited:


Joined Nov 17, 2003

The links to Dick Cappel's website are giving a 404 - not sure if it is a local problem on my end, a problem at Dick's site, or if it is a broken link.

Great work, btw.



Joined May 11, 2009
hey will this circuit stimulate on microcap 9.0....??????????????
if yes then plz send me some information on how to do that..................i am studing cicuit simulation and have to submit a project in college..can u plz help me...!!!!!!!!!!!!!!!i am realy in need of a project related to microcap....\

i wud be heartly thankfull..........!!!!!!
mail me info at <snip>
Last edited by a moderator:


Joined Dec 14, 2011
Dear Jpanhalt:
could you be so kind as to point me to the figures and the circuit diagram of your current prove for osciloscope, for I am not being able to find them.
Thank you a lot


Joined Jan 16, 2014
I got this:-
Internal Server Error

The server encountered an internal error or misconfiguration and was unable to complete your request.
Please contact the server administrator, and inform them of the time the error occurred, and anything you might have done that may have caused the error.
More information about this error may be available in the server error log.
Additionally, a 500 Internal Server Error error was encountered while trying to use an ErrorDocument to handle the request.

Not that I'm surprised, give the time. . .


Joined Apr 27, 2014
Hi, am I doing something silly or are there no schematic files available? I can see them for other projects but not this one. I'd like to build this probe ASAP. Cheers, Terry


Joined Mar 24, 2008
Hi, am I doing something silly or are there no schematic files available? I can see them for other projects but not this one. I'd like to build this probe ASAP. Cheers, Terry
See post #8. DickCapples also visits this site regularly, you can send him a PM.