Rogowski Integrator Performance

Which product is right for your exact requirements
Post Reply
mathestro
User
User
Posts: 2
Joined: Thu Jun 10, 2021 12:44 am

Rogowski Integrator Performance

Post by mathestro »

Dear Community,

I've been working in power quality since 2010 measuring a wide range of machines.

Rogowski coils are very easy to manipulate, but it's very problematic when you need to measure power inverters or any kind of high dI/dt signal.

On a high dI/dt, the rogowski integrator outputs an underdamped signal; so a little 1/4hp motor DOL start wont be recorded without a very underdamped step response. No issues when the same motor is measured with ferromagnetic cores. Some times they are present in middle of a measurement with currents of 2000A; supperposed in this high RMS current an underdamped step response.

The rogowski coils I've been used are similar (if not the same) to TA325 and TA326. They are included in many power quality meters as part of the kit. I know Picotech is another level of quality (I have the Series 2000 and the Software is outstanding!) and I wish this integrators were manufactured by Picotech, and were not the same I can find in the global market.

With power inverters the output current is very dirty, due to the pwm of igbt. My theory is that these very short, but high dI/dt, affects the integrator and makes the waveform an absolutely disaster. Why I'm targeting the integrator, because the power quality meter I have (ELSPEC G4500) didn't always use the external integrator, instead of that, they used to run a code in the DSP to integrate the signal (we never noticed underdamped waveforms then). Due to the high temperature on the chip, maybe due to this high energy demanding routine, it was changed to an external integrator, with the consecuent lose of quality of record.

The external integrator seems to be the current standard, due to the simplicity it offers to the power meter manufacturers. Ferromagnetic split current transformers has very low frequency response, commonly 400Hz in the range of 3000A.

My question is the following: Can the TA325 or TA326 be adjusted to measure power inverters with high frequency carriers and no distorting the signal? Has Picotech an integrator capable of measure a solar power inverter like the ABB PVS-800, again, no distorting the signal?

Please let me know your experience with this kind of sensor and maybe other power quality meters.

Thanks in advance.

mathestro
User
User
Posts: 2
Joined: Thu Jun 10, 2021 12:44 am

Re: Rogowski Integrator Performance

Post by mathestro »

Dear Community,

I've taken some screenshots of the waveforms using the old and new visualization software. I couldn't include the pictures, so I put the whole list at the end of the post using ImgBB.

In the first pictures you'll see the original underdamped step response of a motor start waveforms. Measurements were made with the flexible rogowski coils with external integrator.

Pictures 004 and 005 shows the very early stage of the measured waveforms. I'ts very probable that those transient shapes were related with the front-end amplifier; this power quality analyzer use, for each signal, two operational amplifiers with different gain, in a non inverter configuration. The main goal is to keep the accuracy in the nominal range, and in the presence of high inrush signals. When the signal excedes the nominal, a very low resistance conmutator switches the input of ADC, disconnecting the low range amplifier output and connecting the high one.

Pictures 006 to 008 show the same time frame but with dc harmonic, fundamental and harmonics greater than two. Take a look to the dc harmonic, it has a period of 2055 miliseconds; it can't be produced inside the instrument, I don't know how; it must come from the outside.

Pictures 009 to 012 show another two motor starting waveforms.

Now, picture 013 shows the new visualization software, and its effort to fix this issue.

Picture 014 shows a very sophisticated mechanic to do it. Seems to be an attempt to get a decent RMS values, but the waveform looks awful. Each cycle is carefully relocated to obtain a more real shape. On the early stage a very unreal glitch that touches 180 amperes; you'll se the dc harmonic, fundamental and other harmonics in the pictures 015 to 017.

Picture 018 is just another example. This shapes doesn't appear in the old visualization software, so it seems to be a good trying to fix this issue.

https://ibb.co/ynJ6Rgk
https://ibb.co/CB96s1J
https://ibb.co/CwcpM30
https://ibb.co/YWTVvP9
https://ibb.co/ZxzNyp0
https://ibb.co/LtrMY0k
https://ibb.co/tsWZMZV
https://ibb.co/vsVyKnK
https://ibb.co/gWd24vR
https://ibb.co/2sjhpNw
https://ibb.co/MMFmCMx
https://ibb.co/Wpk1Y3V
https://ibb.co/fMMLtXL
https://ibb.co/X2YBMxV
https://ibb.co/tm4SxVY
https://ibb.co/cgjtkVj
https://ibb.co/Wv3JyL8
https://ibb.co/0hvYPjh

Gerry
PICO STAFF
PICO STAFF
Posts: 970
Joined: Mon Aug 11, 2014 11:14 am

Re: Rogowski Integrator Performance

Post by Gerry »

Hi mathestro,

Sorry for the late reply.

Unfortunately, we have no other Rogowski coils, and no means of adapting the ones that we have. I had a look at the example images that you posted and I couldn't devise a way to differentiate between the desired waveform and the undesired ringing/level changing with any of the processing tools that we have currently (I could use a math channel to marginally differentiate them using a modified integral, but this decayed over time, as you would expect from integration).

I can only suggest that, at this moment in time, for your more specialised requirement, focussing your attentions on the manufacturers in the Power Quality analysis field would be a better option.

Regards,

Gerry
Gerry
Technical Specialist

Post Reply