Rogowski Coil Redeux

Way back when I was building the capacitor bank, I built a inductive current sensor called a rogowski coil so that I could have some instrumentation to see just what kind of voltage and current I was dealing with when I used my capacitor bank. Sadly I wasn’t able to get it to work.

But first a little bit of introduction, the rogowski coil is essentially an inductive current sensor similar to a current transformer, but different in that it is wound on a flexible coilform that allows temporary installation. Additionally the can be tailored to suit a wide range of currents and are ideal for short high intensity transient pulses (such as capacitive discharge). The coil itself when placed around a conductor has a voltage proportional to dI/dt, however this by itself is fairly useless since anyone carrying out an experiment wants to know I over time.

Figure 1 – Graph of data captured from Rogowski coil integrator using a Tektronix 2012B, arbitrary scaling used. Calibration unknown on current data. dV/dt smoothing by moving average over 4 data points.

In order to extract current over time from dI/dt, you must integrate the voltage. This is done using an opamp as an integrator. By itself not terribly challenging; the hard part is data capture/acquisition, and calibration. Note that because the coil itself gives dI/dt, the peak rate of change of the circuit voltage should coincide with the peak current after integration, which is a pretty good indicator of whether you have the right configuration for your integrator.

Originally I had tried to use my sound card with an integrator + buffered circuit, however with the level of AC decoupling present, terribly ineffective with a very slow response. The trick is to avoid extra AC decoupling, and simply correct for voltage offset, which in itself is a good idea, however the challenging part was compensating for the voltage offset. If you don’t correct for voltage offset using opamps, and simply AC decouple, you introduce a RC network at every junction, and blur your current waveform over a longer period of time.

Last week I set up my old original coilgun with the rogowski coil, and a DSO that I have access to and began capturing data to test a circuit, with some success. I finalized the circuit topology and laid out the remaining tasks:

• Prototype rogowski coil integrator with attenuator, integrator, gain stage, and line driver
• Build current shunt reference for DSO
• Build DC amplifier to provide current waveform
• Calibrate stepped attenuator, gain stage to match current reference
• Build final version, calibrate, package in metal enclosure

2.5 kJ Capacitor Bank Complete!

After plenty of delays due to more important projects, I have finally completed the capacitor bank control system and construction.

The capacitor bank is a 22,500uF, 450V capacitor bank with a 10kA surge current capacity using solid state switching. A custom built charger was created so that I could use 12VDC (ie: car batteries) to charge the capacitor bank outdoors for more energetic tests. Along with the charger, a control system to monitor bank voltage, and provide safety interlocks to prevent discharge during firing, automatic voltage control, and discharge load control to abort high-current tests were key aspects of the charge controller.

Cap Bank – Rogowski Coil

So before I really get started with my capacitor bank, I decided I would design on paper all the instrumentation so I could characterize the capacitor bank discharges. In order to measure current over time, either current shunts (DC), current transformers (AC), or rogowski coils (AC, pulse) can be used. The advantage of a rogowski coil is that it is equivalent to a air-core current transformer allowing for much faster response to changes in current flow. Also, due to construction, a rogowski coil can be opened and closed for placement in temporary positions.

Yesterday I finished my constructing my rogowski coil, however it still requires an active integrator to make voltage proportional to current. I used RG6 coax cable, stripped the outer sheet, braid, and foil, then wound 30AWG magnet wire evenly around the dielectric. The magnet wire was attached to braid and core at either end, heat-shrinked, and luer-lock syringe fittings were placed on the ends to allow for the coil to be placed around an object.

A major key to being able to construct a accurate rogowski coil is that the windings must be absolutely even, and remain even as the coil is bent and closed. It is easier to wind a coil on a straight segment of dielectric, heatshrink, then bend, instead of attempting to precision-wind around a torroidal coilform.

The luer-lock fittings prior to attachment.

The completed rogowski coil, note the heatshrink around the coil and BNC connector on the end.

Closeup of the joint where the coil closes. I didn’t have small enough heat-shrink so I had to wedge some tiny pieces of balsa wood in to keep everything snug and secure.

Once I had the coil built I gave it a test, first using a small motor, and a second test using my old coilgun.

I used a induction motor fan wired up to 120VAC for the first test, it didn’t seem to be affected by an un-centered conductor, however I may still build a plexiglass support so that under higher-voltages there is no risk of arcing into the coil (the heatshrink + enamel can only stand so much).

The test setup and waveform (sine = 120VAC, other = current).

The second test was performed using my old 430J coilgun, I looped the coil around the heavy cable from the stud-SCR and tested a few voltages. I captured the waveforms using a soundcard oscilloscope program and got some data, however the higher powers generated voltages exceeding the max rating of the soundcard, causing clipping.

The 3 test waveforms captured at increasing capacitor voltages, note the last test at 430V caused significant clipping of the current waveform.

And also, 2 videos (the last 2 tests)