Suppose I have a current transformer where the primary current is half-wave rectified.
In the ideal (Spice, natch ;-), nothing will show on the secondary.
In real life, what would I see? ...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at
formatting link
| 1962 | I love to cook with wine. Sometimes I even put it in the food.
It sure should. The AC component should couple through.
Most AC power type CTs are laminated steel and will saturate at maybe a per cent or so DC current, which will greatly reduce the secondary signal. And wreck the CT until you demagnetize it.
High frequency CTs could be powder cores, more tolerant of DC.
--
John Larkin Highland Technology, Inc
picosecond timing precision measurement
jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com
It sounds like there's something wrong with your Spice simulation. In a real life that doesn't have the core saturation issues that John pointed out, you should see half-wave rectified current, offset downward so that it's average value is 0.
--
Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
Works just fine if the flux has a chance to return to zero. If you have pulsed DC on the primary, you need a diode in series with the burden resistor to achieve this.
In the ideal case (no saturation), you should see your primary current waveform reflected in the secondary winding. In the real world, you will get accurate secondary waveforms until the core saturates, then you will get a much smaller secondary waveform as the CT essentially is transformed into an air core transformer. Roughly...
The transformer can be self-reset for volt-seconds when the forward current stops flowing, if a secondary-side rectifier is also present to produce a DC output signal. The amplitude of this reset voltage is limited by stray capacitance and voltage breakdown of the output diode.
If secondary is simply terminated with a resistor, you'll still get an AC output signal. The transformer is heavily into the saturated region at the end of the current pulse, but still functions as a transformer. LF bandwidth is reduced (droop)due to the degraded effective permeability.
In the Rogowski extreme case, the permeability is that of free space, and the current waveform is so 'droopy' that it has to be reconstructed by differentiation.
LEM-type trsnsducers, on the other hand, attempt to keep flux in the core zeroed, by matching ampere turns in the secondary.
LEM Current sensors which are focused on Power applications (those ABB too ;-) ) are based on Hall-Effect chips or flux-gate certainly not as simple as a traditional air-gap xformer. e.g. an open-loop current is basically an air-gap laminated steel (as JL has pointed) well ... steel ... not really the core is an alloy FeNi-Si(2%).
BTW femm.info is a great tool for magnetostatic simulation.
There are many ways to skin a cat. The principle of zero flux null for low frequency and DC measurement is achievable in many ways, hence 'LEM-type' and not LEM.
Not sure what you mean by open-loop current.......Current sensors can be discussed quite comprehensively without ever having to refer to core gapping or grades of steel.
Have not had much joy from magnetic simulators and cannot seriously recommend one, for most practical applications. Good to hear you are happy with something in the public domain. I'd be impressed to see anything you might produce, using it, that would address the OP's original inquiry.
Sorry to hear you've had such bad luck. I second the recommendation to use femm 4.2 It's power is awesome. Easy to get better than 1% absolute accuracy. I've used it in designing NDE Instrumentation, to predict Eddy Currents and System performance. It was the only tool I could find that gave me better than 1ppm resolution to improve the electronic design, which had better than 10ppm capabilities.
Plus, the results include skin effect, which enables you to fairly accurately model higher harmonics than just AC mains. The MOST important contribution of femm 4.2 is the way you can manipulate 'proposed' solutions and learn a great deal about magnetics, and skin effect. I used it to disprove MANY an old concept that has been drug around for years and years.
As one example, a colleague in Brazil and I worked together to model a 3 phase power transformer fairly accurately. The final model resulted in components you could put into LTspice and really analyze what's going on in a circuit. Showed the inherent weakness of 'available' physical tranformer structures.
I regularly have used femm 4.2 for analyzing current transformers, not for Jim's application. Especially useful was the ability of femm 4.2 to predict the harmonic waveforms you'd get while 'trying' to monitor SMPS inputs, with and without filtering.
I have tried several. Historically, I started with the Russians' Quick Field, was very satisfied with results, then went on to use femm 4.2, and have even used Ansoft's HFSS for analyzing PCB's and coax transitions. But for speed and control of EXACTLY what's going on I use femm 4.2 almost exclusively now. I've used it even in 1 to 10GHz designs to determine the required pitch for stitching GND vias, find cross talk and PCB comparisons using FR4 vs Rogers materials to justify the extra expense. All in all, very powerful tool.
Open-Loop and close-loop hall effect current sensors.
And you're wrong. Materials are crucial for 1% accuracy over I range and frequency.
No way. NDA with ABB. But you may see
formatting link
that the Flux Density with femm.info 2D simulation (I=50A, F[0..50KHz]) is not uniform through the section of a FeNi or FeSi(2%) material. The advanced job with femm.info was to insert the right air gap for uniformize B. The internal rectangle is copper a bus bar, the upper renctangle is FeNi.
You're right there is nothing here to adresse precisely JT inquiry :( ... Never know ...
I was just looking for a simple-minded way to detect the difference in current flow, full-wave AC versus half-wave rectified, no measurement needed, just the state.
...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: skypeanalog | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
I love to cook with wine. Sometimes I even put it in the food.
Drawing the line between inherent and 'by design' could be made easier with quick and cheap simulation. That IS the goal.
A major part of most designs is trying to make functional parts with available materials. Usually you're battling with some shape that has no redeaming virtues save for it's 'low profile' or multiple-sourcing.
What relevence did CT precision have in that kind of comparative situation? Or were you looking for basic application suitability? I hope the femm SW wasn't your reference...... RL
Don't know if an half wave could saturate the core with time ... i think if the hysteresis B(H) cycles are not overlapped exactly then chances are that the core should saturate, may be.
For zeroing the core (Bremanent=0) we applied something like sin(x)/x or something like that, very efficient.
An intentionally undersized saturable AC sensor, before rectification might make the most 'sense'. If it's full wave, you'd get healthy pulses at the reversal frequency. If there's no reversal, output amplitude would plummet.
If you use the diode reset version on the rectified DC line, it's output would degrade when reset volt-seconds wasn't present, giving a signal only after each missing half phase.
Well, draw the BH curve and note the core will be magnetized half-way; secondary will be pulses according to the amount of magnetization decrease from nominal towards zero. Expect operation similar to core memory sense wire. May help to have another "DC reset" line to set average bias near zero. With large drive, sense line (secondary) may only be pulses.
I spoke with a fellow who worked at current transformer plant. He brought up one thing they did that nobody could explain, but had to be done or the transformers dodn't work.
They'd wind the toroids then loops a few turns of wire outside the core and across it, in a way that would make no sense at all. He even asked what the point was, nobody knew anymore but that's just what they did and had been doing for years.
Might this have been some type of demagnetization trick?
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.