If you are dealing with REAL CTs, (utility sized, used for equipment protection and revenue metering) the manufacturer will supply saturation curves, which chart the resistive vs reactive burden and identify the area within which the specified accuracy will be achieved.
Some applications make use of the non-linearities. For example, NEMA motor protective relays (melting alloy and bi-metallic)are available in three standard trip speeds, class 10 (fast) through class 30 (slow). It is common practice to implement the class 30 curve (in large sizes, anyway) by using a class 20 (standard) trip element with CTs pushed into their non-linear range.
Old textbooks seem to consider that the largest source of error will be from I-quadrature, dismissing I-inphase as almost a second order. I suspect that this may be because the burdens of that day would most likely be meters, which could be quite inductive.
AFAICS if the load is known to be resistive then errors in the current ratio will be mainly due to I-inphase, from the core loss.
I did have a chance to test this theory about 10-15 years ago, with a contract to investigate the cause of mfr'ing spread in precision 400Hz/500Arms CT's... was it due to turns-count or core problems (Ni-Fe cores).
Cores from various batches were carefully hand-wound with 1000-turn secondaries, and I made up a jig (out of tapped brass spacers) that centred a wound CT on an 8-turn primary. I had a milliWattmeter and access to a 100Arms precision current source, and ammeter, both with traceable calibration. Overall accuracy was estimated to be of the order of 0.03%.
We stimulated the secondaries at various voltages and measured the watts-loss. Then took the CTs up to the high current jig and measured the current-ratio.
There was good correlation between the current-ratio errors and the inphase currents calculated from the watts-loss at the same (internal) secondary voltage.
In fact the problem turned out to be a wider than promised spread in core loss, instead of 3-8mW some cores were up to 15mW (at 5Vrms/400Hz Vsec).
Unfortunately no, because watts-loss is a non-linear function with respect to the secondary voltage.
Probably academic anyway Fred, because I suspect that the OP should be measuring RMS, rather than averaging. Might do better to just go out and buy a LEM/HEME Hall Effect device, the version with a dc output proportional to the RMS of the primary current.
I've already got a rule that ignores threads where the root message is posted via Google (with exceptions for a few notable people)
--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)
Right, it takes a Omron K3NX-VA1C or K3NX-AA1C to do the job without a rectifier. Your scheme should work fine, because the CT load is after the rectifier.
Yo Ho. If you had read the specifications for your 882gb you would know that all you need or want is just the CT. The process meter has facilities for AC current input.
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.