ACS730 500kHz Oscilloscope Current Probe

by undersilicon in Circuits > Electronics

6934 Views, 39 Favorites, 0 Comments

ACS730 500kHz Oscilloscope Current Probe

Portada.jpg

If you've ever worked with some sort of power electronics or high current devices before, you may have also asked yourself how the current waveform looked in those devices, specially if you were troubleshooting them.

Well, since this is not an unusual topic, there are already a number of commercial solutions. They are called current probes and allow engineers to see and measure current waveforms in an oscilloscope up to tens or hundreds of amps.

Without these devices, it would be very diffficult to work in the power applications field, but they are pricey, though. In fact, the starting price for one of these has surely prevented a lot of people to do it, specially in the hobbyist scene.

There are a huge variety of current probes in the market, everyone built with a different approach in mind. Most of them allow to measure current in a non-invasive way relying in custom magnetic transducers, which stands as the main reason for this things being so expensive.

While this is a very desirable feature, it is very difficult to replicate for the regular maker, since there are no commercially available resources to do it. Therefore, the described current probe in this article uses two terminals and a high bandwidth current transducer IC working in the same way as a multimeter.

This allows for an affordable design which could be easily replicated by the average maker.

Features

ACS730-Typical-Application.png

The probe is designed around an ACS730 IC from Allegro MicroSystems, which features a -3dB bandwidth of 1 MHz and +-50 A of measurable range. It is effectively a galvanically isolated hall effect current transducer in a SOIC8 package. This allows for measuring AC mains applications up to 297 VRMS or 420 pk-VDC.

Next comes a signal conditioning stage, built with a pair of AD823ARZ precision high speed operational amplifiers, selected to meet the bandwidth specification of the current sensor. They scale the signal to 100 mV/A.

The probe receives power from a 1000 mAh LiPo battery charged at 200 mAh via Mini USB with an MCP73831 IC. The positive and negative voltages that the probe requires, come from a TPS65133 power converter. The power consumption rises to about 50 mA, achieving a continuous operation of 15 - 20 hours.

In a nutshell, the probe specifications can be summarized as follows:

- Scale: 100 mV/A

- +-25 A of nominal measurable range (40 Apeak)

- Nominal output voltage: +-2.5 V (+-4 Vmax)

- Bandwidth: DC - 500 kHz -1 dB (1 MHz -3dB)

- 15 - 20 hours of operation.

Theory of Operation

SCH.PNG

As stated in the previous step, the main component which makes the probe tick is the ACS730 (U2), outputting 2.5V when the current is zero. Then, if a current is applied at its terminals, the output varies at a rate of 40 mV/A in its +-50 A version. This gives us a voltage between 0.5 V and 4.5 V depending on the current polarity.

As you may have already noticed, there is always a DC offset of 2.5 V imposed in the measurement signal. So, the next section is an AD823ARZ operational amplifier that buffers 2.5 V obtained through a voltage divider from the 5 V supply rail. Using a dedicated voltage reference will result in a much better noise performance but at the expense of increasing the overall cost of the instrument.

Then, a second stage of operational amplifiers remove the offset subtracting the 2.5 VDC buffered level from the output signal (IC2A). The last stage applies a gain of 2.5 to obtain the scale of 100 mV/A (IC2B).

In order to avoid saturation of the operational amplifiers, the probe should not be used with currents above 40 A. Due to limitations in the PCB substrate, 25 A seems like a more reasonable maximum current, although higher transient currents could still be measurable.

The rest of the schematic is mostly self-explanatory. A TPS65133 IC (U3), delivers positive and negative 5 V power rails to the operational amplifiers and current sensor. It is a power converter switching in the megahertz range to maintain good regulation and noise under control.

A 3.7V 1000 mAh LiPo battery provides power to the instrument. An MCP73831 (U1) keeps the charging of said battery at 200 mAh when the MiniUSB (J1) port is connected. An LED lights up when charge is complete.

Performance and Iimitations

1616605710472.jpg
1616605710367.jpg
1616605710423.jpg
Respuesta.PNG

The ACS730 offers a -3dB bandwidth of 1 MHz and a rise time of 0.6 microseconds, which is great, but for a current probe, an attenuated waveform is not really trustworthy and distorts measures. Thus, I suggest using the probe to test signals under the -1db bandwidth of 500 kHz.

This is not a problem for the operational amplifiers stage. As you can see in the pictures, a 1 MHz input sinewave suffers no attenuation or distortion at all. This is well beyond the specs of the current transducer.

Experimental Results

1616691684535.jpg
1616694879504.jpg
1616691684719.jpg
1616691684745.jpg
1616691684616.jpg
1616691684591.jpg
1616691684642.jpg
Scope 1A.bmp
Scope 5A.bmp

I have performed a series of accuracy tests, in which the probe performed beautifully. At the moment, I don't have any AC current source available to show the actual bandwidth of the probe, but I tested the overall DC performance with a regular bench PSU.

When the power supply is set to 1 A of DC current, at the same time, a current clamp is measuring 0.96A of actual current circulating through the wires, and a multimeter is showing 0.0995 V at the output of the probe, which is exactly as it should be. Applying 5 A, leads to even better results, with an actual current of 4.86 A in the clamp, the multimeter is reading 0.4855 V at the output. That is an impressive accuracy figure for a 30€ device!

Oscilloscope tests show a fair bit of noise coming from the probe, although it is significant at low level currents (less than 1 A), it is less concerning at higher amplitude currents. This could be due to improper testing of the signal with an oscilloscope probe with a long ground lead.

Project Files and Documentation

Caja_2.PNG
Caja_1.PNG
Captura_cab.PNG
PCB_top.PNG

The archive below contains all the information to replicate this project.

CAD files (Fusion 360 and EAGLE 9.X):

- F360 assembly project (Assembly.f3z)

- F360 electronics project (ACS730_probe.f3z)

- F360 Schematic (F360_ACS730_sch.fsch)

- F360 PCB layout (F360_ACS730_PCB.fbrd)

- EAGLE Schematic (EAGLE_ACS730_sch.sch)

- EAGLE PCB layout (EAGLE_ACS730_PCB.brd)

- STL models of the featured box (3DModels.zip)

Printable files:

- Schematic (ACS730_Probe_Schematic.pdf)

Production files:

- Bill of materials (BOM_ACS730_25A.txt, BOM_ACS730_25A.csv)

- Gerber files (ACS730_25A.zip)

This project was made entirely with Fusion 360. Electrical files of the project have been exported as Fusion 360 and EAGLE 9.X compatible files. Google Drive link to access the project files:

https://drive.google.com/file/d/14_Hp3iuTNa1wZ3X5k...

I hope you liked this project! As it is my first instructabe I encourage you to comment and share this article, so I could write better ones in the future. I would be glad to answer all of your questions.

Enjoy!

You can also follow me on Instagram!