Archived Forum

Which ct are you using, which input circuit? (EmonTx - which, Arduino, or what?)

Two 2.5 V zeners WON'T clamp it to 5 V - think what the reverse characteristic of a zener is. But depending on the input circuitry, it might not be necessary and almost certainly won't be the best way.

The analogue input has protection diodes built in, and these have a maximum current rating of 40 mA, so it is the current, not the voltage, that you need to limit. You need to calculate or measure the peak voltage you get from your ct (see my report on the YHDC one to see what the shape of the burden voltage might be) and use a series resistor between the burden and the input to limit that current. That resistor is already present in the emonTx V3. The maximum recommended value according to the Atmel data sheet is 10 kΩ, which should be more than enough without having a significant effect on the accuracy.

Which CT circuit?

Well, a homebrew actually. I'm using 40A CR8448-2500-N CT's feeding into multiple ADS1015 i2c converters running at 5v.

The ADS1015 specifically states that no voltage over VDD+0.3 should be applied to any pin.

Built around the building blocks documentation on this site. Using 10k voltage divider with 10uF ceramic bypass and 110 ohm burden resistor (4.97Vpp).

Sorry for throwing a curve ball, but I appreciate the help.

Hmm, just came back from learning about the drawbacks of using zeners in this type of application.

Plan B:

Purchase and design around a 200A CT to ensure I don't exceed the input voltage then use the adjustable gain of the ADC to increase resolution once the startup surge has subsided.

Unless of course there are other suggestions ?

I've got to come back to what I wrote above. Just look at the data sheet, page 9 - the inputs are shown with protection diodes!  And you didn't read this bit properly either: "To prevent the ESD diodes from turning on, the absolute voltage on any input must stay within the following range:  GND – 0.3V < AINx < VDD + 0.3V"

That "GND – 0.3V" and "V_DD + 0.3 V" are entirely reasonable since a voltage source is expected to be able to supply infinite current. Hence anything more than 0.3 V outside the supply rails will source an unacceptably high current into the diode. The data sheet tells you what the acceptable limit is, in Absolute Maximum Ratings.  As the input impedance is in the megohm region, there is of course only one way to get that sort of input current. So now it should be very obvious, especially when you read all of page 9, what you need to do.

I think that if you limit the current to, say, 2-3 mA, you will be perfectly safe.

And don't expect to see 18 V p-p out of the ct either, since even though it is revenue grade, it will be starting to saturate at 3.75 x rated current.

Gee, can it be as simple as adding a (1.7k which I have plenty of) resistor in series with the input to limit the current and let the internal diodes do their thing ?

There's an old saying that goes something like this: "Engineers can do for pennies what everybody else does for pounds." [Insert the appropriate units for your currency.]   I'm an engineer.

Your help is very much appreciated. My interpretation of datasheet subtleties has also improved with your guidance.

Thank you.