Current transformer

That's why engineers shouldn't have managers.

OTOH engineers are too often prone to fail to appreciate other aspects of a business like marketing, pointless gimmicks etc.

NT

The biggest problem with engineers (assuming technical competence) (excepting that most can't spell) is that they can tend to work on the wrong problem, specifically things that are most interesting and least necessary. Engineers would often rather spend a month inventing a new, super-efficient hammer, instead of half an hour pounding nails.

Marketing people are not much better at estimating a market than investors are good at predicting the stock market. The dynamics are similar. Most (I speak through expensive experience) marketing people want to market more than they want to sell.

So the prime function of management is to make sure everybody is working toward developing things that will actually sell and pay the bills.

i.e. too much or too little of something is bad

-Lasse

The appropriate SI unit of measure is the onosecond.

Dare to be stupid! You learn more that way.

If the BIPM had known about that one a bit earlier, they'd have avoided a few real genuine floaters. One such was standardizing the candela, which is expressible in kilograms, metres, and seconds times a very squishy standard curve for subjective brightness vs. wavelength that gets changed periodically.

Cheers

Phil Hobbs

Interesting subject Sylvia, thanks for posting. I was just reading up on this and found a possible reason for the 220R: When attaching the probe to a live wire, without burden resistor rather high voltages will be created. The diodes may not be able to handle this. Also, If the diodes die, a person touching the un-burdened wires will get zapped. What you think?

Tony

But not enough, if John Larkin is the example ...

A Model-T cost $400, too. So?

It could be belt and braces, yes. Though I'm left wondering why that resistance is so high. On my estimation, based on the very limited testing I've done, the voltage with just a 220 ohm burden, and at the maximum rated current, would be in the region of 50V peak to peak. That's just about within the safe range, I suppose, but a lower resistance would make it safer.

Sylvia.

The best managers I've had are good at shielding their employees from their managers. I tell my current boss that he's a great s*it-umbrella. ...and I usually only see him once or twice a week.

Just the diodes would probably be more reliable and fail-safe, as they normally fail shorted, while resistors more often fail open. The diodes are probably 1N400x which handle 1 amp. The 50V P-P is about 15 VRMS on the 220 ohm, or about 68 mA, barely tickling the diodes, but over one watt in the resistor.

I have been looking at some PCB mounted CTs, both AC and Hall Effect. These clamp on AC types have 3000:1 ratio with 30A and 100A ratings, for $16:

This 15A Hall effect sensor is only $12:

And I have one of these 100A donut type sensors that was only $9:

From a discussion about a Fluke clamp-on adapter, on the

It may be interesting to discuss ways of measuring DC current using non-contact methods. The standard method is a Hall-effect device:

This article explains a lot about AC and DC current sensing, but mostly Hall Effect:

There is another method, using the magnetoresistive effect, that is gaining popularity:

But even that requires electronics, and a power source.

I think there is another method, which operates on the principle of a saturable core reactor, where the measured current saturates the toroidal core, and its inductance is measured by the instrument. It may be possible to introduce a variable DC current of opposite polarity to the measured current, and use some method to sense when the fields cancel.

The only way to determine the DC current flow in a conductor while maintaining isoolation is to measure the external magnetic field, or flux. Here is some interesting info:

Maybe too much information, and still doesn't explain the weird circuit of this CT.

Paul

There are some cool, hyper-accurate DC current transformers. Bergoz, Danfysic, Lem.

(Which explains the DCCT concept. I did one once, and it was a nuisance.)

It looks like Danfysic has discontinued theirs.

That was uncalled for. Please grow up.

Jeroen Belleman

...and it was a single-sided floppy (ie: before DOS 1.1).

Some requests are just never rewarded with the desired response.

NT

Current measurement is something I'm much interested in. Unfortunately all but 1 of your links are broken (freepatentsonline).

NT

Oops! I copy/pasted from my post in the forum, and the displayed URLs were shortened.

Here are the full links:

($16 clamp-on AC, 30A, 100A)

($12 AC/DC 15A)

($9 AC/DC 100A)

The standard method is a Hall-effect device:

This article explains a lot about AC and DC current sensing, but mostly Hall Effect:

There is another method, using the magnetoresistive effect, that is gaining popularity:

The only way to determine the DC current flow in a conductor while maintaining isolation is to measure the external magnetic field, or flux. Here is some interesting info:

HTH :)

Paul